基于Java实现的OpenCV封装开源项目

QianXun (QianXun) • 2025年10月15日 02:58

                        <!DOCTYPE html>
<html lang="zh-CN">
<head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>基于Java实现的OpenCV封装开源项目：原理、架构与设计思想</title>
    <link href="https://fonts.googleapis.com/icon?family=Material+Icons" rel="stylesheet">
    <link href="https://fonts.googleapis.com/css2?family=Noto+Sans+SC:wght@400;500;700&family=Roboto+Mono:wght@400;500&display=swap" rel="stylesheet">
    <style>
        :root {
            --primary-color: #1a237e;
            --primary-light: #534bae;
            --primary-dark: #000051;
            --secondary-color: #2962ff;
            --text-on-primary: #ffffff;
            --text-primary: #212121;
            --text-secondary: #757575;
            --background: #f5f7fa;
            --card-bg: #ffffff;
            --code-bg: #263238;
            --code-color: #aed581;
            --border-color: #e0e0e0;
            --accent-color: #ff4081;
        }

        * {
            margin: 0;
            padding: 0;
            box-sizing: border-box;
        }

        body {
            font-family: 'Noto Sans SC', sans-serif;
            background-color: var(--background);
            color: var(--text-primary);
            line-height: 1.6;
        }

        .poster-container {
            width: 960px;
            min-height: 2000px;
            margin: 0 auto;
            padding: 40px 20px;
            background: linear-gradient(135deg, #f5f7fa 0%, #e4e8f0 100%);
            position: relative;
            overflow: visible;
        }

        .poster-header {
            text-align: center;
            margin-bottom: 40px;
            padding: 30px 0;
            background: linear-gradient(135deg, var(--primary-color) 0%, var(--primary-light) 100%);
            border-radius: 12px;
            color: var(--text-on-primary);
            box-shadow: 0 8px 16px rgba(0, 0, 0, 0.1);
        }

        .poster-title {
            font-size: 42px;
            font-weight: 700;
            margin-bottom: 16px;
            letter-spacing: -0.5px;
        }

        .poster-subtitle {
            font-size: 20px;
            font-weight: 400;
            opacity: 0.9;
            max-width: 80%;
            margin: 0 auto;
        }

        .section {
            margin-bottom: 40px;
            background-color: var(--card-bg);
            border-radius: 12px;
            padding: 30px;
            box-shadow: 0 4px 12px rgba(0, 0, 0, 0.05);
        }

        .section-title {
            font-size: 28px;
            font-weight: 700;
            color: var(--primary-color);
            margin-bottom: 20px;
            display: flex;
            align-items: center;
        }

        .section-title .material-icons {
            margin-right: 12px;
            color: var(--secondary-color);
        }

        .subsection {
            margin-bottom: 25px;
        }

        .subsection-title {
            font-size: 22px;
            font-weight: 500;
            color: var(--primary-dark);
            margin-bottom: 15px;
            border-left: 4px solid var(--secondary-color);
            padding-left: 12px;
        }

        p {
            font-size: 16px;
            margin-bottom: 16px;
            color: var(--text-primary);
        }

        ul, ol {
            padding-left: 24px;
            margin-bottom: 16px;
        }

        li {
            margin-bottom: 8px;
            font-size: 16px;
        }

        .highlight {
            background-color: rgba(41, 98, 255, 0.1);
            padding: 2px 6px;
            border-radius: 4px;
            font-weight: 500;
        }

        .code-block {
            background-color: var(--code-bg);
            border-radius: 8px;
            padding: 20px;
            margin: 20px 0;
            overflow-x: auto;
            box-shadow: 0 4px 8px rgba(0, 0, 0, 0.1);
        }

        .code-block pre {
            color: var(--code-color);
            font-family: 'Roboto Mono', monospace;
            font-size: 14px;
            line-height: 1.5;
            white-space: pre-wrap;
            word-wrap: break-word;
        }

        .code-comment {
            color: #608b4e;
        }

        .code-keyword {
            color: #569cd6;
        }

        .code-string {
            color: #ce9178;
        }

        .code-function {
            color: #dcdcaa;
        }

        .architecture-diagram {
            display: flex;
            flex-direction: column;
            align-items: center;
            margin: 30px 0;
        }

        .diagram-layer {
            width: 80%;
            padding: 15px;
            margin: 8px 0;
            border-radius: 8px;
            text-align: center;
            font-weight: 500;
            box-shadow: 0 2px 6px rgba(0, 0, 0, 0.1);
        }

        .layer-java {
            background-color: rgba(41, 98, 255, 0.2);
            border: 2px solid var(--secondary-color);
        }

        .layer-wrapper {
            background-color: rgba(26, 35, 126, 0.15);
            border: 2px solid var(--primary-color);
        }

        .layer-native {
            background-color: rgba(0, 0, 81, 0.1);
            border: 2px solid var(--primary-dark);
        }

        .comparison-table {
            width: 100%;
            border-collapse: collapse;
            margin: 20px 0;
            font-size: 15px;
        }

        .comparison-table th, .comparison-table td {
            padding: 12px 15px;
            text-align: left;
            border-bottom: 1px solid var(--border-color);
        }

        .comparison-table th {
            background-color: rgba(41, 98, 255, 0.1);
            font-weight: 500;
            color: var(--primary-dark);
        }

        .comparison-table tr:nth-child(even) {
            background-color: rgba(0, 0, 0, 0.02);
        }

        .feature-card {
            background-color: var(--card-bg);
            border-radius: 8px;
            padding: 20px;
            margin-bottom: 16px;
            box-shadow: 0 2px 8px rgba(0, 0, 0, 0.05);
            border-left: 4px solid var(--secondary-color);
        }

        .feature-title {
            font-size: 18px;
            font-weight: 500;
            color: var(--primary-dark);
            margin-bottom: 10px;
            display: flex;
            align-items: center;
        }

        .feature-title .material-icons {
            font-size: 20px;
            margin-right: 8px;
            color: var(--secondary-color);
        }

        .footer {
            text-align: center;
            margin-top: 40px;
            padding: 20px;
            color: var(--text-secondary);
            font-size: 14px;
        }

        .decorative-circle {
            position: absolute;
            border-radius: 50%;
            opacity: 0.1;
            z-index: -1;
        }

        .circle-1 {
            width: 300px;
            height: 300px;
            background: linear-gradient(135deg, var(--secondary-color) 0%, var(--primary-light) 100%);
            top: -100px;
            right: -100px;
        }

        .circle-2 {
            width: 200px;
            height: 200px;
            background: linear-gradient(135deg, var(--accent-color) 0%, var(--secondary-color) 100%);
            bottom: 300px;
            left: -80px;
        }
    </style>
</head>
<body>
    <div class="poster-container">
        <div class="decorative-circle circle-1"></div>
        <div class="decorative-circle circle-2"></div>
        
        <header class="poster-header">
            <h1 class="poster-title">基于Java实现的OpenCV封装开源项目</h1>
            <p class="poster-subtitle">原理、架构与设计思想深度解析</p>
        </header>

        <section class="section">
            <h2 class="section-title">
                <i class="material-icons">info</i>
                引言：OpenCV与Java的结合
            </h2>
            <p>OpenCV（开源计算机视觉库）是一个跨平台的计算机视觉和机器学习软件库，提供了丰富的图像处理和计算机视觉算法。最初以C++实现，现已扩展支持Python、Java等多种编程语言。Java作为企业级应用开发的主流语言，与OpenCV的结合为开发者提供了在Java生态中应用计算机视觉技术的可能。</p>
            <p>Java与OpenCV的结合主要通过封装技术实现，即将原生C++代码封装为Java可调用的接口。这种封装不仅保留了OpenCV的高性能特性，还充分利用了Java的平台无关性和丰富的生态系统，使得开发者能够在Java应用中轻松集成计算机视觉功能。</p>
        </section>

        <section class="section">
            <h2 class="section-title">
                <i class="material-icons">category</i>
                主要Java OpenCV封装项目概述
            </h2>
            
            <div class="subsection">
                <h3 class="subsection-title">JavaCV</h3>
                <p>JavaCV是基于OpenCV和其他计算机视觉/多媒体库的Java接口，提供了Java平台上的高效计算机视觉和音视频处理能力。它封装了多个底层C/C++库（如OpenCV、FFmpeg、libdc1394等），并通过Java Native Access(JNA)或JavaCPP技术实现Java调用。</p>
                <div class="feature-card">
                    <div class="feature-title">
                        <i class="material-icons">stars</i>
                        核心特点
                    </div>
                    <ul>
                        <li><span class="highlight">多库集成</span>：不仅封装了OpenCV，还整合了FFmpeg、libdc1394、ARToolKitPlus等多个强大的多媒体库</li>
                        <li><span class="highlight">跨平台支持</span>：支持Windows、Linux、macOS、Android等平台，提供预编译的本地库</li>
                        <li><span class="highlight">高性能</span>：通过JavaCPP直接调用本地代码，减少JNI开发复杂度，支持GPU加速</li>
                        <li><span class="highlight">易用性</span>：提供Java风格的API，兼容Java生态（如Maven/Gradle依赖管理）</li>
                    </ul>
                </div>
            </div>
            
            <div class="subsection">
                <h3 class="subsection-title">OpenCV官方Java绑定</h3>
                <p>OpenCV本身也提供了官方的Java绑定，直接在OpenCV源码中通过JNI技术实现。这种绑定方式更加贴近原生OpenCV API，能够及时跟进OpenCV的更新。</p>
                <div class="feature-card">
                    <div class="feature-title">
                        <i class="material-icons">stars</i>
                        核心特点
                    </div>
                    <ul>
                        <li><span class="highlight">官方支持</span>：由OpenCV团队直接维护，与OpenCV核心库同步更新</li>
                        <li><span class="highlight">API一致性</span>：Java API与C++ API保持高度一致，便于参考官方文档</li>
                        <li><span class="highlight">性能优化</span>：直接通过JNI调用，减少了中间层的开销</li>
                        <li><span class="highlight">平台支持</span>：支持主流操作系统，但需要手动配置本地库路径</li>
                    </ul>
                </div>
            </div>
            
            <div class="subsection">
                <h3 class="subsection-title">其他封装项目</h3>
                <p>除了上述两个主要项目外，还有一些其他的Java OpenCV封装实现，如通过JNI或JNA技术封装OpenCV的项目。这些项目通常针对特定需求或场景进行了优化，如简化API、增强特定功能等。</p>
            </div>
        </section>

        <section class="section">
            <h2 class="section-title">
                <i class="material-icons">architecture</i>
                封装原理与架构
            </h2>
            
            <div class="subsection">
                <h3 class="subsection-title">JNI技术</h3>
                <p>Java Native Interface（JNI）是Java平台的标准机制，允许Java代码调用本地应用程序和库（如C/C++）。OpenCV官方Java绑定就是基于JNI实现的。</p>
                <div class="code-block">
                    <pre><span class="code-comment">// JNI调用示例</span>
<span class="code-keyword">public class</span> <span class="code-function">OpenCVNative</span> {
    <span class="code-keyword">static</span> {
        System.loadLibrary(Core.NATIVE_LIBRARY_NAME);
    }
    
    <span class="code-comment">// 声明本地方法</span>
    <span class="code-keyword">public native</span> <span class="code-keyword">long</span> <span class="code-function">n_Mat</span>();
    <span class="code-keyword">public native</span> <span class="code-keyword">void</span> <span class="code-function">n_delete</span>(<span class="code-keyword">long</span> addr);
}</pre>
                </div>
            </div>
            
            <div class="subsection">
                <h3 class="subsection-title">JNA技术</h3>
                <p>Java Native Access（JNA）是另一种Java调用本地库的技术，相比JNI更加简单，不需要编写C/C++代码。JavaCV主要使用JNA或JavaCPP（基于JNA的增强版）来实现封装。</p>
                <div class="code-block">
                    <pre><span class="code-comment">// JNA调用示例</span>
<span class="code-keyword">public interface</span> <span class="code-function">OpenCVLibrary</span> <span class="code-keyword">extends</span> Library {
    OpenCVLibrary INSTANCE = (OpenCVLibrary) Native.load(<span class="code-string">"opencv_core"</span>, OpenCVLibrary.<span class="code-keyword">class</span>);
    
    <span class="code-comment">// 直接映射C函数</span>
    Pointer cvCreateImage(int width, int height, int depth, int channels);
    <span class="code-keyword">void</span> cvReleaseImage(Pointer image);
}</pre>
                </div>
            </div>
            
            <div class="subsection">
                <h3 class="subsection-title">架构层次</h3>
                <div class="architecture-diagram">
                    <div class="diagram-layer layer-java">Java应用层</div>
                    <div class="diagram-layer layer-wrapper">OpenCV封装层 (JavaCV/OpenCV Java绑定)</div>
                    <div class="diagram-layer layer-wrapper">JNI/JNA桥接层</div>
                    <div class="diagram-layer layer-native">OpenCV原生库 (C++)</div>
                </div>
                <p>这种分层架构使得Java开发者能够使用熟悉的Java API，同时利用OpenCV的高性能原生实现。封装层负责将Java调用转换为对原生库的调用，并处理数据类型转换、内存管理等复杂问题。</p>
            </div>
        </section>

        <section class="section">
            <h2 class="section-title">
                <i class="material-icons">lightbulb</i>
                设计思想与最佳实践
            </h2>
            
            <div class="subsection">
                <h3 class="subsection-title">面向对象设计模式的应用</h3>
                <p>Java OpenCV封装项目广泛应用了多种设计模式，以提高代码的可维护性和可扩展性：</p>
                <ul>
                    <li><span class="highlight">工厂模式</span>：用于创建不同类型的图像处理对象，如FrameGrabber、FrameRecorder等</li>
                    <li><span class="highlight">适配器模式</span>：将OpenCV的C++接口适配为Java风格的API</li>
                    <li><span class="highlight">外观模式</span>：提供简化的高级接口，隐藏底层复杂性</li>
                    <li><span class="highlight">单例模式</span>：管理全局资源，如OpenCV库的加载和初始化</li>
                </ul>
            </div>
            
            <div class="subsection">
                <h3 class="subsection-title">内存管理策略</h3>
                <p>由于Java和C++的内存管理机制不同，封装项目需要特别关注内存管理问题：</p>
                <ul>
                    <li><span class="highlight">资源自动释放</span>：实现AutoCloseable接口，支持try-with-resources语法</li>
                    <li><span class="highlight">引用计数</span>：跟踪对象引用，防止过早释放内存</li>
                    <li><span class="highlight">弱引用</span>：使用弱引用避免内存泄漏</li>
                    <li><span class="highlight">内存池</span>：重用频繁分配的内存块，减少GC压力</li>
                </ul>
            </div>
            
            <div class="subsection">
                <h3 class="subsection-title">API设计原则</h3>
                <p>优秀的Java OpenCV封装项目遵循以下API设计原则：</p>
                <ul>
                    <li><span class="highlight">一致性</span>：API命名和结构保持一致，降低学习成本</li>
                    <li><span class="highlight">简洁性</span>：提供简洁的高级API，同时保留底层访问能力</li>
                    <li><span class="highlight">可扩展性</span>：设计灵活的接口，便于添加新功能</li>
                    <li><span class="highlight">Java风格</span>：遵循Java编程习惯，如异常处理、集合使用等</li>
                </ul>
            </div>
        </section>

        <section class="section">
            <h2 class="section-title">
                <i class="material-icons">code</i>
                代码示例
            </h2>
            
            <div class="subsection">
                <h3 class="subsection-title">使用JavaCV进行图像处理</h3>
                <div class="code-block">
                    <pre><span class="code-keyword">import</span> org.bytedeco.javacv.*;
<span class="code-keyword">import</span> org.bytedeco.opencv.opencv_core.*;
<span class="code-keyword">import static</span> org.bytedeco.opencv.global.opencv_imgcodecs.*;
<span class="code-keyword">import static</span> org.bytedeco.opencv.global.opencv_imgproc.*;

<span class="code-keyword">public class</span> <span class="code-function">ImageProcessingExample</span> {
    <span class="code-keyword">public static void</span> main(String[] args) {
        <span class="code-comment">// 加载图像</span>
        Mat image = imread(<span class="code-string">"input.jpg"</span>);
        
        <span class="code-comment">// 创建灰度图像</span>
        Mat grayImage = <span class="code-keyword">new</span> Mat();
        cvtColor(image, grayImage, COLOR_BGR2GRAY);
        
        <span class="code-comment">// 边缘检测</span>
        Mat edges = <span class="code-keyword">new</span> Mat();
        Canny(grayImage, edges, <span class="code-number">100</span>, <span class="code-number">200</span>);
        
        <span class="code-comment">// 保存结果</span>
        imwrite(<span class="code-string">"edges.jpg"</span>, edges);
        
        <span class="code-comment">// 显示结果</span>
        CanvasFrame canvas = <span class="code-keyword">new</span> CanvasFrame(<span class="code-string">"Edge Detection"</span>);
        canvas.showImage(edges);
        canvas.waitKey();
    }
}</pre>
                </div>
            </div>
            
            <div class="subsection">
                <h3 class="subsection-title">使用OpenCV Java绑定进行人脸检测</h3>
                <div class="code-block">
                    <pre><span class="code-keyword">import</span> org.opencv.core.*;
<span class="code-keyword">import</span> org.opencv.imgcodecs.Imgcodecs;
<span class="code-keyword">import</span> org.opencv.objdetect.CascadeClassifier;

<span class="code-keyword">public class</span> <span class="code-function">FaceDetectionExample</span> {
    <span class="code-keyword">public static void</span> main(String[] args) {
        <span class="code-comment">// 加载OpenCV库</span>
        System.loadLibrary(Core.NATIVE_LIBRARY_NAME);
        
        <span class="code-comment">// 加载人脸检测级联分类器</span>
        CascadeClassifier faceDetector = <span class="code-keyword">new</span> CascadeClassifier();
        faceDetector.load(<span class="code-string">"haarcascade_frontalface_alt.xml"</span>);
        
        <span class="code-comment">// 读取图像</span>
        Mat image = Imgcodecs.imread(<span class="code-string">"input.jpg"</span>);
        
        <span class="code-comment">// 检测人脸</span>
        MatOfRect faceDetections = <span class="code-keyword">new</span> MatOfRect();
        faceDetector.detectMultiScale(image, faceDetections);
        
        <span class="code-comment">// 在图像上标记人脸</span>
        <span class="code-keyword">for</span> (Rect rect : faceDetections.toArray()) {
            Imgproc.rectangle(
                image,                          <span class="code-comment">// 目标图像</span>
                <span class="code-keyword">new</span> Point(rect.x, rect.y),      <span class="code-comment">// 矩形左上角</span>
                <span class="code-keyword">new</span> Point(rect.x + rect.width, rect.y + rect.height), <span class="code-comment">// 矩形右下角</span>
                <span class="code-keyword">new</span> Scalar(<span class="code-number">0</span>, <span class="code-number">255</span>, <span class="code-number">0</span>),     <span class="code-comment">// 颜色 (绿色)</span>
                <span class="code-number">3</span>                               <span class="code-comment">// 线宽</span>
            );
        }
        
        <span class="code-comment">// 保存结果</span>
        Imgcodecs.imwrite(<span class="code-string">"faces_detected.jpg"</span>, image);
    }
}</pre>
                </div>
            </div>
        </section>

        <section class="section">
            <h2 class="section-title">
                <i class="material-icons">trending_up</i>
                应用场景与未来发展方向
            </h2>
            
            <div class="subsection">
                <h3 class="subsection-title">主要应用场景</h3>
                <div class="feature-card">
                    <div class="feature-title">
                        <i class="material-icons">security</i>
                        安防监控
                    </div>
                    <p>人脸识别、行为分析、异常检测等，广泛应用于智慧城市、智能安防系统</p>
                </div>
                
                <div class="feature-card">
                    <div class="feature-title">
                        <i class="material-icons">directions_car</i>
                        智能驾驶
                    </div>
                    <p>车道检测、障碍物识别、交通标志识别等，为自动驾驶系统提供视觉感知能力</p>
                </div>
                
                <div class="feature-card">
                    <div class="feature-title">
                        <i class="material-icons">shopping_cart</i>
                        零售与电商
                    </div>
                    <p>商品识别、客流分析、虚拟试衣等，提升购物体验和运营效率</p>
                </div>
                
                <div class="feature-card">
                    <div class="feature-title">
                        <i class="material-icons">medical_services</i>
                        医疗影像
                    </div>
                    <p>医学图像分析、病灶检测、影像增强等，辅助医生进行诊断</p>
                </div>
            </div>
            
            <div class="subsection">
                <h3 class="subsection-title">未来发展方向</h3>
                <ul>
                    <li><span class="highlight">深度学习集成</span>：更紧密地集成TensorFlow、PyTorch等深度学习框架，支持模型加载和推理</li>
                    <li><span class="highlight">GPU加速</span>：增强对CUDA、OpenCL等GPU计算框架的支持，提升处理性能</li>
                    <li><span class="highlight">云原生支持</span>：适应容器化、微服务架构，提供云原生计算机视觉服务</li>
                    <li><span class="highlight">跨平台一致性</span>：进一步统一不同平台间的API和行为，简化跨平台开发</li>
                    <li><span class="highlight">实时流处理</span>：增强对实时视频流处理的支持，满足低延迟应用需求</li>
                </ul>
            </div>
            
            <div class="subsection">
                <h3 class="subsection-title">JavaCV与OpenCV Java绑定对比</h3>
                <table class="comparison-table">
                    <tr>
                        <th>特性</th>
                        <th>JavaCV</th>
                        <th>OpenCV Java绑定</th>
                    </tr>
                    <tr>
                        <td>封装技术</td>
                        <td>JNA/JavaCPP</td>
                        <td>JNI</td>
                    </tr>
                    <tr>
                        <td>API风格</td>
                        <td>Java风格，更符合Java习惯</td>
                        <td>接近C++ API，更贴近OpenCV原生</td>
                    </tr>
                    <tr>
                        <td>功能范围</td>
                        <td>多库集成，不仅限于OpenCV</td>
                        <td>专注于OpenCV功能</td>
                    </tr>
                    <tr>
                        <td>更新频率</td>
                        <td>独立更新，可能滞后于OpenCV</td>
                        <td>与OpenCV同步更新</td>
                    </tr>
                    <tr>
                        <td>性能</td>
                        <td>略低于JNI，但差距很小</td>
                        <td>最高，直接调用原生代码</td>
                    </tr>
                    <tr>
                        <td>易用性</td>
                        <td>较高，Maven依赖管理简单</td>
                        <td>中等，需手动配置本地库</td>
                    </tr>
                </table>
            </div>
        </section>

        <footer class="footer">
            <p>© 2023 基于Java实现的OpenCV封装开源项目研究 | 技术分享海报</p>
        </footer>
    </div>
</body>
</html>                    

讨论回复

1 条回复

QianXun (QianXun) #1

10-17 07:00

                                        <!DOCTYPE html>
<html lang="zh-CN">
<head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>JavaCV详尽教程</title>
    <link href="https://fonts.googleapis.com/icon?family=Material+Icons" rel="stylesheet">
    <link href="https://fonts.googleapis.com/css2?family=Noto+Sans+SC:wght@400;500;700&family=Roboto+Mono:wght@400;500&display=swap" rel="stylesheet">
    <style>
        :root {
            --primary-color: #1a237e;
            --primary-light: #534bae;
            --primary-dark: #000051;
            --secondary-color: #2962ff;
            --text-on-primary: #ffffff;
            --text-primary: #212121;
            --text-secondary: #757575;
            --background: #f5f7fa;
            --card-bg: #ffffff;
            --code-bg: #263238;
            --code-color: #aed581;
            --border-color: #e0e0e0;
            --accent-color: #ff4081;
        }

        * {
            margin: 0;
            padding: 0;
            box-sizing: border-box;
        }

        body {
            font-family: 'Noto Sans SC', sans-serif;
            background-color: var(--background);
            color: var(--text-primary);
            line-height: 1.6;
        }

        .poster-container {
            width: 960px;
            min-height: 2000px;
            margin: 0 auto;
            padding: 40px 20px;
            background: linear-gradient(135deg, #f5f7fa 0%, #e4e8f0 100%);
            position: relative;
            overflow: visible;
        }

        .poster-header {
            text-align: center;
            margin-bottom: 40px;
            padding: 30px 0;
            background: linear-gradient(135deg, var(--primary-color) 0%, var(--primary-light) 100%);
            border-radius: 12px;
            color: var(--text-on-primary);
            box-shadow: 0 8px 16px rgba(0, 0, 0, 0.1);
        }

        .poster-title {
            font-size: 42px;
            font-weight: 700;
            margin-bottom: 16px;
            letter-spacing: -0.5px;
        }

        .poster-subtitle {
            font-size: 20px;
            font-weight: 400;
            opacity: 0.9;
            max-width: 80%;
            margin: 0 auto;
        }

        .section {
            margin-bottom: 40px;
            background-color: var(--card-bg);
            border-radius: 12px;
            padding: 30px;
            box-shadow: 0 4px 12px rgba(0, 0, 0, 0.05);
        }

        .section-title {
            font-size: 28px;
            font-weight: 700;
            color: var(--primary-color);
            margin-bottom: 20px;
            display: flex;
            align-items: center;
        }

        .section-title .material-icons {
            margin-right: 12px;
            color: var(--secondary-color);
        }

        .subsection {
            margin-bottom: 25px;
        }

        .subsection-title {
            font-size: 22px;
            font-weight: 500;
            color: var(--primary-dark);
            margin-bottom: 15px;
            border-left: 4px solid var(--secondary-color);
            padding-left: 12px;
        }

        p {
            font-size: 16px;
            margin-bottom: 16px;
            color: var(--text-primary);
        }

        ul, ol {
            padding-left: 24px;
            margin-bottom: 16px;
        }

        li {
            margin-bottom: 8px;
            font-size: 16px;
        }

        .highlight {
            background-color: rgba(41, 98, 255, 0.1);
            padding: 2px 6px;
            border-radius: 4px;
            font-weight: 500;
        }

        .code-block {
            background-color: var(--code-bg);
            border-radius: 8px;
            padding: 20px;
            margin: 20px 0;
            overflow-x: auto;
            box-shadow: 0 4px 8px rgba(0, 0, 0, 0.1);
        }

        .code-block pre {
            color: var(--code-color);
            font-family: 'Roboto Mono', monospace;
            font-size: 14px;
            line-height: 1.5;
            white-space: pre-wrap;
            word-wrap: break-word;
        }

        .code-comment {
            color: #608b4e;
        }

        .code-keyword {
            color: #569cd6;
        }

        .code-string {
            color: #ce9178;
        }

        .code-function {
            color: #dcdcaa;
        }

        .feature-card {
            background-color: var(--card-bg);
            border-radius: 8px;
            padding: 20px;
            margin-bottom: 16px;
            box-shadow: 0 2px 8px rgba(0, 0, 0, 0.05);
            border-left: 4px solid var(--secondary-color);
        }

        .feature-title {
            font-size: 18px;
            font-weight: 500;
            color: var(--primary-dark);
            margin-bottom: 10px;
            display: flex;
            align-items: center;
        }

        .feature-title .material-icons {
            font-size: 20px;
            margin-right: 8px;
            color: var(--secondary-color);
        }

        .step-container {
            display: flex;
            margin-bottom: 20px;
        }

        .step-number {
            flex: 0 0 40px;
            height: 40px;
            background-color: var(--secondary-color);
            color: white;
            border-radius: 50%;
            display: flex;
            align-items: center;
            justify-content: center;
            font-weight: 700;
            margin-right: 16px;
        }

        .step-content {
            flex: 1;
        }

        .api-table {
            width: 100%;
            border-collapse: collapse;
            margin: 20px 0;
            font-size: 15px;
        }

        .api-table th, .api-table td {
            padding: 12px 15px;
            text-align: left;
            border-bottom: 1px solid var(--border-color);
        }

        .api-table th {
            background-color: rgba(41, 98, 255, 0.1);
            font-weight: 500;
            color: var(--primary-dark);
        }

        .api-table tr:nth-child(even) {
            background-color: rgba(0, 0, 0, 0.02);
        }

        .decorative-circle {
            position: absolute;
            border-radius: 50%;
            opacity: 0.1;
            z-index: -1;
        }

        .circle-1 {
            width: 300px;
            height: 300px;
            background: linear-gradient(135deg, var(--secondary-color) 0%, var(--primary-light) 100%);
            top: -100px;
            right: -100px;
        }

        .circle-2 {
            width: 200px;
            height: 200px;
            background: linear-gradient(135deg, var(--accent-color) 0%, var(--secondary-color) 100%);
            bottom: 300px;
            left: -80px;
        }

        .footer {
            text-align: center;
            margin-top: 40px;
            padding: 20px;
            color: var(--text-secondary);
            font-size: 14px;
        }
    </style>
</head>
<body>
    <div class="poster-container">
        <div class="decorative-circle circle-1"></div>
        <div class="decorative-circle circle-2"></div>
        
        <header class="poster-header">
            <h1 class="poster-title">JavaCV详尽教程</h1>
            <p class="poster-subtitle">从入门到精通的Java计算机视觉开发指南</p>
        </header>

        <section class="section">
            <h2 class="section-title">
                <i class="material-icons">download</i>
                安装与配置
            </h2>
            
            <div class="subsection">
                <h3 class="subsection-title">Maven依赖配置</h3>
                <p>JavaCV通过Maven中央仓库提供，只需在pom.xml中添加以下依赖：</p>
                <div class="code-block">
                    <pre><span class="code-comment">&lt;!-- JavaCV核心库 --&gt;</span>
&lt;dependency&gt;
    &lt;groupId&gt;org.bytedeco&lt;/groupId&gt;
    &lt;artifactId&gt;javacv&lt;/artifactId&gt;
    &lt;version&gt;1.5.7&lt;/version&gt;
&lt;/dependency&gt;

<span class="code-comment">&lt;!-- 根据需要添加特定平台依赖 --&gt;</span>
&lt;dependency&gt;
    &lt;groupId&gt;org.bytedeco&lt;/groupId&gt;
    &lt;artifactId&gt;opencv&lt;/artifactId&gt;
    &lt;version&gt;4.5.5-1.5.7&lt;/version&gt;
    &lt;classifier&gt;windows-x86_64&lt;/classifier&gt; <span class="code-comment">// 根据平台选择</span>
&lt;/dependency&gt;</pre>
                </div>
            </div>
            
            <div class="subsection">
                <h3 class="subsection-title">平台特定依赖</h3>
                <p>JavaCV支持多种平台，需要根据目标平台选择相应的预编译库：</p>
                <ul>
                    <li><span class="highlight">Windows</span>: windows-x86_64, windows-x86</li>
                    <li><span class="highlight">Linux</span>: linux-x86_64, linux-armhf, linux-arm64</li>
                    <li><span class="highlight">macOS</span>: macosx-x86_64, macosx-arm64</li>
                    <li><span class="highlight">Android</span>: android-arm, android-x86</li>
                </ul>
            </div>
            
            <div class="subsection">
                <h3 class="subsection-title">IDE配置</h3>
                <div class="step-container">
                    <div class="step-number">1</div>
                    <div class="step-content">
                        <p>在IntelliJ IDEA或Eclipse中创建新的Maven项目</p>
                    </div>
                </div>
                <div class="step-container">
                    <div class="step-number">2</div>
                    <div class="step-content">
                        <p>添加上述Maven依赖到pom.xml文件</p>
                    </div>
                </div>
                <div class="step-container">
                    <div class="step-number">3</div>
                    <div class="step-content">
                        <p>刷新项目依赖，确保所有库正确下载</p>
                    </div>
                </div>
                <div class="step-container">
                    <div class="step-number">4</div>
                    <div class="step-content">
                        <p>设置JVM参数（可选）：-Djava.library.path=/path/to/native/libs</p>
                    </div>
                </div>
            </div>
        </section>

        <section class="section">
            <h2 class="section-title">
                <i class="material-icons">api</i>
                核心API介绍
            </h2>
            
            <div class="subsection">
                <h3 class="subsection-title">主要组件</h3>
                <table class="api-table">
                    <tr>
                        <th>组件</th>
                        <th>功能描述</th>
                        <th>主要用途</th>
                    </tr>
                    <tr>
                        <td>Frame</td>
                        <td>表示图像或视频帧的基本数据结构</td>
                        <td>存储图像数据、时间戳等元信息</td>
                    </tr>
                    <tr>
                        <td>FrameGrabber</td>
                        <td>用于从各种源捕获帧的抽象类</td>
                        <td>从摄像头、视频文件、图像序列获取帧</td>
                    </tr>
                    <tr>
                        <td>FrameRecorder</td>
                        <td>用于将帧写入各种目标的抽象类</td>
                        <td>将帧保存为视频文件、图像序列</td>
                    </tr>
                    <tr>
                        <td>CanvasFrame</td>
                        <td>用于显示图像帧的GUI组件</td>
                        <td>实时预览、调试显示</td>
                    </tr>
                    <tr>
                        <td>OpenCVFrameConverter</td>
                        <td>在Frame和OpenCV Mat之间转换</td>
                        <td>与OpenCV原生API互操作</td>
                    </tr>
                </table>
            </div>
            
            <div class="subsection">
                <h3 class="subsection-title">FrameGrabber实现类</h3>
                <ul>
                    <li><span class="highlight">OpenCVFrameGrabber</span>：使用OpenCV后端，支持多种视频格式和摄像头</li>
                    <li><span class="highlight">FFmpegFrameGrabber</span>：基于FFmpeg，支持更广泛的视频格式和网络流</li>
                    <li><span class="highlight">DC1394FrameGrabber</span>：专用于IEEE 1394工业相机</li>
                    <li><span class="highlight">FlyCaptureFrameGrabber</span>：用于Point Grey工业相机</li>
                    <li><span class="highlight">OpenKinectFrameGrabber</span>：用于Kinect深度相机</li>
                </ul>
            </div>
        </section>

        <section class="section">
            <h2 class="section-title">
                <i class="material-icons">image</i>
                图像处理示例
            </h2>
            
            <div class="subsection">
                <h3 class="subsection-title">基本图像操作</h3>
                <div class="code-block">
                    <pre><span class="code-keyword">import</span> org.bytedeco.javacv.*;
<span class="code-keyword">import</span> org.bytedeco.opencv.opencv_core.*;
<span class="code-keyword">import static</span> org.bytedeco.opencv.global.opencv_imgcodecs.*;
<span class="code-keyword">import static</span> org.bytedeco.opencv.global.opencv_imgproc.*;

<span class="code-keyword">public class</span> <span class="code-function">BasicImageOperations</span> {
    <span class="code-keyword">public static void</span> main(String[] args) <span class="code-keyword">throws</span> Exception {
        <span class="code-comment">// 1. 加载图像</span>
        OpenCVFrameConverter.ToMat converter = <span class="code-keyword">new</span> OpenCVFrameConverter.ToMat();
        Mat image = imread(<span class="code-string">"input.jpg"</span>);
        
        <span class="code-comment">// 2. 转换为灰度图像</span>
        Mat grayImage = <span class="code-keyword">new</span> Mat();
        cvtColor(image, grayImage, COLOR_BGR2GRAY);
        
        <span class="code-comment">// 3. 图像缩放</span>
        Mat resizedImage = <span class="code-keyword">new</span> Mat();
        Size size = <span class="code-keyword">new</span> Size(image.cols() / 2, image.rows() / 2);
        resize(image, resizedImage, size);
        
        <span class="code-comment">// 4. 图像旋转</span>
        Mat rotatedImage = <span class="code-keyword">new</span> Mat();
        Point center = <span class="code-keyword">new</span> Point(image.cols() / 2, image.rows() / 2);
        Mat rotationMatrix = getRotationMatrix2D(center, <span class="code-number">45</span>, <span class="code-number">1.0</span>);
        warpAffine(image, rotatedImage, rotationMatrix, image.size());
        
        <span class="code-comment">// 5. 边缘检测</span>
        Mat edges = <span class="code-keyword">new</span> Mat();
        Canny(grayImage, edges, <span class="code-number">100</span>, <span class="code-number">200</span>);
        
        <span class="code-comment">// 6. 保存结果</span>
        imwrite(<span class="code-string">"gray.jpg"</span>, grayImage);
        imwrite(<span class="code-string">"resized.jpg"</span>, resizedImage);
        imwrite(<span class="code-string">"rotated.jpg"</span>, rotatedImage);
        imwrite(<span class="code-string">"edges.jpg"</span>, edges);
        
        <span class="code-comment">// 7. 显示结果</span>
        CanvasFrame canvas = <span class="code-keyword">new</span> CanvasFrame(<span class="code-string">"Image Processing"</span>, <span class="code-number">1</span>);
        canvas.setDefaultCloseOperation(javax.swing.JFrame.EXIT_ON_CLOSE);
        
        <span class="code-comment">// 显示原始图像</span>
        canvas.showImage(converter.convert(image));
        canvas.waitKey(<span class="code-number">2000</span>);
        
        <span class="code-comment">// 显示边缘检测结果</span>
        canvas.showImage(converter.convert(edges));
        canvas.waitKey();
    }
}</pre>
                </div>
            </div>
            
            <div class="subsection">
                <h3 class="subsection-title">图像滤波与增强</h3>
                <div class="code-block">
                    <pre><span class="code-keyword">import</span> org.bytedeco.opencv.opencv_core.*;
<span class="code-keyword">import static</span> org.bytedeco.opencv.global.opencv_imgproc.*;

<span class="code-comment">// 高斯模糊</span>
Mat blurred = <span class="code-keyword">new</span> Mat();
GaussianBlur(image, blurred, <span class="code-keyword">new</span> Size(<span class="code-number">15</span>, <span class="code-number">15</span>), <span class="code-number">0</span>);

<span class="code-comment">// 双边滤波（保边去噪）</span>
Mat bilateral = <span class="code-keyword">new</span> Mat();
bilateralFilter(image, bilateral, <span class="code-number">15</span>, <span class="code-number">80</span>, <span class="code-number">80</span>);

<span class="code-comment">// 亮度与对比度调整</span>
Mat adjusted = <span class="code-keyword">new</span> Mat();
image.convertTo(adjusted, CV_8UC3, <span class="code-number">1.5</span>, <span class="code-number">30</span>); <span class="code-comment">// 对比度1.5，亮度+30</span>

<span class="code-comment">// 直方图均衡化（增强对比度）</span>
Mat equalized = <span class="code-keyword">new</span> Mat();
equalizeHist(grayImage, equalized);

<span class="code-comment">// 形态学操作</span>
Mat morphed = <span class="code-keyword">new</span> Mat();
Mat kernel = getStructuringElement(MORPH_RECT, <span class="code-keyword">new</span> Size(<span class="code-number">5</span>, <span class="code-number">5</span>));
morphologyEx(image, morphed, MORPH_CLOSE, kernel);</pre>
                </div>
            </div>
        </section>

        <section class="section">
            <h2 class="section-title">
                <i class="material-icons">videocam</i>
                视频处理示例
            </h2>
            
            <div class="subsection">
                <h3 class="subsection-title">视频读取与处理</h3>
                <div class="code-block">
                    <pre><span class="code-keyword">import</span> org.bytedeco.javacv.*;
<span class="code-keyword">import</span> org.bytedeco.opencv.opencv_core.*;
<span class="code-keyword">import static</span> org.bytedeco.opencv.global.opencv_imgproc.*;

<span class="code-keyword">public class</span> <span class="code-function">VideoProcessing</span> {
    <span class="code-keyword">public static void</span> main(String[] args) <span class="code-keyword">throws</span> Exception {
        <span class="code-comment">// 1. 创建视频抓取器</span>
        FFmpegFrameGrabber grabber = <span class="code-keyword">new</span> FFmpegFrameGrabber(<span class="code-string">"input.mp4"</span>);
        grabber.start();
        
        <span class="code-comment">// 2. 获取视频信息</span>
        <span class="code-keyword">int</span> width = grabber.getImageWidth();
        <span class="code-keyword">int</span> height = grabber.getImageHeight();
        <span class="code-keyword">double</span> frameRate = grabber.getVideoFrameRate();
        <span class="code-keyword">int</span> frameCount = grabber.getLengthInFrames();
        
        System.out.println(<span class="code-string">"视频尺寸: "</span> + width + <span class="code-string">"x"</span> + height);
        System.out.println(<span class="code-string">"帧率: "</span> + frameRate);
        System.out.println(<span class="code-string">"总帧数: "</span> + frameCount);
        
        <span class="code-comment">// 3. 创建视频录制器</span>
        FFmpegFrameRecorder recorder = <span class="code-keyword">new</span> FFmpegFrameRecorder(<span class="code-string">"output.mp4"</span>, width, height);
        recorder.setVideoCodec(avcodec.AV_CODEC_ID_H264);
        recorder.setFormat(<span class="code-string">"mp4"</span>);
        recorder.setFrameRate(frameRate);
        recorder.setPixelFormat(avutil.AV_PIX_FMT_YUV420P);
        recorder.start();
        
        <span class="code-comment">// 4. 创建转换器</span>
        OpenCVFrameConverter.ToMat converter = <span class="code-keyword">new</span> OpenCVFrameConverter.ToMat();
        
        <span class="code-comment">// 5. 处理每一帧</span>
        Frame frame;
        <span class="code-keyword">int</span> processedFrames = <span class="code-number">0</span>;
        
        <span class="code-keyword">while</span> ((frame = grabber.grab()) != <span class="code-keyword">null</span>) {
            <span class="code-comment">// 转换为OpenCV Mat</span>
            Mat mat = converter.convert(frame);
            
            <span class="code-comment">// 处理帧（例如：边缘检测）</span>
            Mat gray = <span class="code-keyword">new</span> Mat();
            cvtColor(mat, gray, COLOR_BGR2GRAY);
            
            Mat edges = <span class="code-keyword">new</span> Mat();
            Canny(gray, edges, <span class="code-number">100</span>, <span class="code-number">200</span>);
            
            <span class="code-comment">// 转换回彩色</span>
            Mat result = <span class="code-keyword">new</span> Mat();
            cvtColor(edges, result, COLOR_GRAY2BGR);
            
            <span class="code-comment">// 记录处理后的帧</span>
            recorder.record(converter.convert(result));
            
            processedFrames++;
            
            <span class="code-comment">// 打印进度</span>
            <span class="code-keyword">if</span> (processedFrames % <span class="code-number">100</span> == <span class="code-number">0</span>) {
                System.out.println(<span class="code-string">"已处理: "</span> + processedFrames + <span class="code-string">"/"</span> + frameCount + <span class="code-string">" 帧"</span>);
            }
        }
        
        <span class="code-comment">// 6. 释放资源</span>
        grabber.stop();
        recorder.stop();
        
        System.out.println(<span class="code-string">"视频处理完成！"</span>);
    }
}</pre>
                </div>
            </div>
            
            <div class="subsection">
                <h3 class="subsection-title">实时摄像头处理</h3>
                <div class="code-block">
                    <pre><span class="code-keyword">public class</span> <span class="code-function">CameraProcessing</span> {
    <span class="code-keyword">public static void</span> main(String[] args) <span class="code-keyword">throws</span> Exception {
        <span class="code-comment">// 1. 创建摄像头抓取器</span>
        OpenCVFrameGrabber grabber = <span class="code-keyword">new</span> OpenCVFrameGrabber(<span class="code-number">0</span>); <span class="code-comment">// 0表示默认摄像头</span>
        grabber.setImageWidth(<span class="code-number">640</span>);
        grabber.setImageHeight(<span class="code-number">480</span>);
        grabber.start();
        
        <span class="code-comment">// 2. 创建显示窗口</span>
        CanvasFrame canvas = <span class="code-keyword">new</span> CanvasFrame(<span class="code-string">"Camera Feed"</span>, <span class="code-number">1</span>);
        canvas.setDefaultCloseOperation(javax.swing.JFrame.EXIT_ON_CLOSE);
        
        <span class="code-comment">// 3. 创建转换器</span>
        OpenCVFrameConverter.ToMat converter = <span class="code-keyword">new</span> OpenCVFrameConverter.ToMat();
        
        <span class="code-comment">// 4. 实时处理循环</span>
        Frame frame;
        <span class="code-keyword">while</span> ((frame = grabber.grab()) != <span class="code-keyword">null</span>) {
            <span class="code-comment">// 转换为OpenCV Mat</span>
            Mat mat = converter.convert(frame);
            
            <span class="code-comment">// 应用滤镜（例如：卡通效果）</span>
            Mat result = applyCartoonEffect(mat);
            
            <span class="code-comment">// 显示处理后的帧</span>
            canvas.showImage(converter.convert(result));
            
            <span class="code-comment">// 检查窗口是否关闭</span>
            <span class="code-keyword">if</span> (canvas.isVisible() == <span class="code-keyword">false</span>) {
                <span class="code-keyword">break</span>;
            }
        }
        
        <span class="code-comment">// 5. 释放资源</span>
        grabber.stop();
        canvas.dispose();
    }
    
    <span class="code-keyword">private static</span> Mat applyCartoonEffect(Mat src) {
        <span class="code-comment">// 卡通效果实现</span>
        Mat imgGray = <span class="code-keyword">new</span> Mat();
        cvtColor(src, imgGray, COLOR_BGR2GRAY);
        
        Mat imgBlur = <span class="code-keyword">new</span> Mat();
        medianBlur(imgGray, imgBlur, <span class="code-number">7</span>);
        
        Mat imgEdge = <span class="code-keyword">new</span> Mat();
        adaptiveThreshold(imgBlur, imgEdge, <span class="code-number">255</span>, ADAPTIVE_THRESH_MEAN_C, THRESH_BINARY, <span class="code-number">7</span>, <span class="code-number">7</span>);
        
        Mat imgColor = <span class="code-keyword">new</span> Mat();
        bilateralFilter(src, imgColor, <span class="code-number">15</span>, <span class="code-number">80</span>, <span class="code-number">80</span>);
        
        Mat cartoon = <span class="code-keyword">new</span> Mat();
        bitwise_and(imgColor, imgColor, cartoon, imgEdge);
        
        <span class="code-keyword">return</span> cartoon;
    }
}</pre>
                </div>
            </div>
        </section>

        <section class="section">
            <h2 class="section-title">
                <i class="material-icons">auto_awesome</i>
                高级功能应用
            </h2>
            
            <div class="subsection">
                <h3 class="subsection-title">人脸检测与识别</h3>
                <div class="code-block">
                    <pre><span class="code-keyword">import</span> org.bytedeco.javacv.*;
<span class="code-keyword">import</span> org.bytedeco.opencv.opencv_core.*;
<span class="code-keyword">import</span> org.bytedeco.opencv.opencv_objdetect.*;
<span class="code-keyword">import static</span> org.bytedeco.opencv.global.opencv_imgproc.*;

<span class="code-keyword">public class</span> <span class="code-function">FaceDetection</span> {
    <span class="code-keyword">public static void</span> main(String[] args) <span class="code-keyword">throws</span> Exception {
        <span class="code-comment">// 1. 加载人脸检测器</span>
        CascadeClassifier faceDetector = <span class="code-keyword">new</span> CascadeClassifier();
        faceDetector.load(<span class="code-string">"haarcascade_frontalface_alt.xml"</span>);
        
        <span class="code-comment">// 2. 打开摄像头</span>
        OpenCVFrameGrabber grabber = <span class="code-keyword">new</span> OpenCVFrameGrabber(<span class="code-number">0</span>);
        grabber.start();
        
        <span class="code-comment">// 3. 创建显示窗口</span>
        CanvasFrame canvas = <span class="code-keyword">new</span> CanvasFrame(<span class="code-string">"Face Detection"</span>, <span class="code-number">1</span>);
        canvas.setDefaultCloseOperation(javax.swing.JFrame.EXIT_ON_CLOSE);
        
        <span class="code-comment">// 4. 创建转换器</span>
        OpenCVFrameConverter.ToMat converter = <span class="code-keyword">new</span> OpenCVFrameConverter.ToMat();
        
        <span class="code-comment">// 5. 实时人脸检测循环</span>
        Frame frame;
        <span class="code-keyword">while</span> ((frame = grabber.grab()) != <span class="code-keyword">null</span>) {
            <span class="code-comment">// 转换为OpenCV Mat</span>
            Mat mat = converter.convert(frame);
            
            <span class="code-comment">// 转换为灰度图像（人脸检测通常在灰度图像上进行）</span>
            Mat gray = <span class="code-keyword">new</span> Mat();
            cvtColor(mat, gray, COLOR_BGR2GRAY);
            
            <span class="code-comment">// 直方图均衡化（提高检测效果）</span>
            equalizeHist(gray, gray);
            
            <span class="code-comment">// 检测人脸</span>
            RectVector faces = <span class="code-keyword">new</span> RectVector();
            faceDetector.detectMultiScale(gray, faces);
            
            <span class="code-comment">// 在图像上标记人脸</span>
            <span class="code-keyword">for</span> (<span class="code-keyword">int</span> i = <span class="code-number">0</span>; i < faces.size(); i++) {
                Rect face = faces.get(i);
                
                <span class="code-comment">// 绘制人脸矩形框</span>
                rectangle(mat, 
                    <span class="code-keyword">new</span> Point(face.x(), face.y()),
                    <span class="code-keyword">new</span> Point(face.x() + face.width(), face.y() + face.height()),
                    <span class="code-keyword">new</span> Scalar(<span class="code-number">0</span>, <span class="code-number">255</span>, <span class="code-number">0</span>), <span class="code-number">2</span>);
                
                <span class="code-comment">// 在人脸上方添加标签</span>
                putText(mat, <span class="code-string">"Face"</span>, 
                    <span class="code-keyword">new</span> Point(face.x(), face.y() - <span class="code-number">10</span>),
                    FONT_HERSHEY_SIMPLEX, <span class="code-number">0.7</span>, 
                    <span class="code-keyword">new</span> Scalar(<span class="code-number">0</span>, <span class="code-number">255</span>, <span class="code-number">0</span>), <span class="code-number">2</span>);
            }
            
            <span class="code-comment">// 显示结果</span>
            canvas.showImage(converter.convert(mat));
            
            <span class="code-comment">// 检查窗口是否关闭</span>
            <span class="code-keyword">if</span> (canvas.isVisible() == <span class="code-keyword">false</span>) {
                <span class="code-keyword">break</span>;
            }
        }
        
        <span class="code-comment">// 6. 释放资源</span>
        grabber.stop();
        canvas.dispose();
    }
}</pre>
                </div>
            </div>
            
            <div class="subsection">
                <h3 class="subsection-title">对象跟踪</h3>
                <div class="code-block">
                    <pre><span class="code-keyword">import</span> org.bytedeco.javacv.*;
<span class="code-keyword">import</span> org.bytedeco.opencv.opencv_core.*;
<span class="code-keyword">import</span> org.bytedeco.opencv.opencv_videoio.*;
<span class="code-keyword">import org.bytedeco.opencv.opencv_video.*;</span>
<span class="code-keyword">import static</span> org.bytedeco.opencv.global.opencv_imgproc.*;

<span class="code-keyword">public class</span> <span class="code-function">ObjectTracking</span> {
    <span class="code-keyword">public static void</span> main(String[] args) <span class="code-keyword">throws</span> Exception {
        <span class="code-comment">// 1. 打开视频或摄像头</span>
        OpenCVFrameGrabber grabber = <span class="code-keyword">new</span> OpenCVFrameGrabber(<span class="code-string">"input.mp4"</span>);
        grabber.start();
        
        <span class="code-comment">// 2. 创建显示窗口</span>
        CanvasFrame canvas = <span class="code-keyword">new</span> CanvasFrame(<span class="code-string">"Object Tracking"</span>, <span class="code-number">1</span>);
        canvas.setDefaultCloseOperation(javax.swing.JFrame.EXIT_ON_CLOSE);
        
        <span class="code-comment">// 3. 创建转换器</span>
        OpenCVFrameConverter.ToMat converter = <span class="code-keyword">new</span> OpenCVFrameConverter.ToMat();
        
        <span class="code-comment">// 4. 读取第一帧并选择跟踪对象</span>
        Frame frame = grabber.grab();
        Mat firstFrame = converter.convert(frame);
        
        <span class="code-comment">// 让用户选择跟踪区域（这里简化为固定区域）</span>
        Rect roi = <span class="code-keyword">new</span> Rect(<span class="code-number">300</span>, <span class="code-number">200</span>, <span class="code-number">100</span>, <span class="code-number">100</span>);
        rectangle(firstFrame, roi.tl(), roi.br(), <span class="code-keyword">new</span> Scalar(<span class="code-number">0</span>, <span class="code-number">255</span>, <span class="code-number">0</span>), <span class="code-number">2</span>);
        canvas.showImage(converter.convert(firstFrame));
        canvas.waitKey(<span class="code-number">1000</span>);
        
        <span class="code-comment">// 5. 初始化跟踪器</span>
        Tracker tracker = TrackerKCF.create();
        tracker.init(firstFrame, roi);
        
        <span class="code-comment">// 6. 跟踪循环</span>
        <span class="code-keyword">while</span> ((frame = grabber.grab()) != <span class="code-keyword">null</span>) {
            Mat mat = converter.convert(frame);
            
            <span class="code-comment">// 更新跟踪器</span>
            <span class="code-keyword">boolean</span> success = tracker.update(mat, roi);
            
            <span class="code-comment">// 绘制跟踪结果</span>
            <span class="code-keyword">if</span> (success) {
                rectangle(mat, roi.tl(), roi.br(), <span class="code-keyword">new</span> Scalar(<span class="code-number">0</span>, <span class="code-number">255</span>, <span class="code-number">0</span>), <span class="code-number">2</span>);
                putText(mat, <span class="code-string">"Tracking"</span>, roi.tl(), FONT_HERSHEY_SIMPLEX, <span class="code-number">0.7</span>, 
                    <span class="code-keyword">new</span> Scalar(<span class="code-number">0</span>, <span class="code-number">255</span>, <span class="code-number">0</span>), <span class="code-number">2</span>);
            } <span class="code-keyword">else</span> {
                putText(mat, <span class="code-string">"Lost"</span>, <span class="code-keyword">new</span> Point(<span class="code-number">100</span>, <span class="code-number">80</span>), FONT_HERSHEY_SIMPLEX, <span class="code-number">0.7</span>, 
                    <span class="code-keyword">new</span> Scalar(<span class="code-number">0</span>, <span class="code-number">0</span>, <span class="code-number">255</span>), <span class="code-number">2</span>);
            }
            
            <span class="code-comment">// 显示结果</span>
            canvas.showImage(converter.convert(mat));
            
            <span class="code-comment">// 检查窗口是否关闭</span>
            <span class="code-keyword">if</span> (canvas.isVisible() == <span class="code-keyword">false</span>) {
                <span class="code-keyword">break</span>;
            }
        }
        
        <span class="code-comment">// 7. 释放资源</span>
        grabber.stop();
        canvas.dispose();
    }
}</pre>
                </div>
            </div>
        </section>

        <footer class="footer">
            <p>© 2023 JavaCV详尽教程 | 从入门到精通的Java计算机视觉开发指南</p>
        </footer>
    </div>
</body>
</html>                                    

需要登录才能发表回复

登录注册

基于Java实现的OpenCV封装开源项目

讨论回复

推荐

MindSearch: 模拟人类思维的人工智能搜索框架 思·索 — 通过多智能体框架实现深度网络信息搜索与整合

LightRAG: 简单快速的检索增强生成

all-MiniLM-L6-v2模型全面解析

DeepDive系统技术实现与架构分析：基于因子图概率推理的知识抽取框架

🌌 技术栈的星际航路：从组织学视角解码阿里、腾讯、字节与B站的选择

MindSearch: 模拟人类思维的人工智能搜索框架思·索 — 通过多智能体框架实现深度网络信息搜索与整合