OpenCV4.5.5 各模块说明:
提示:主要介绍OpenCV4.5.5中的各个模块,尤其是在Extra modules中,新版本相较于2.x,3.x的版本新增的很多内容,不过很多模块在官方的版本包中是不带的,需要自己源码编译
例如:
- Main modules:
- core. Core functionality 核心基础模块
- Basic structures 基础结构:主要包括OpenCV的数据类型等
- C structures and operations C语言结构和操作
- Operations on arrays 数组操作:加减缩放,求非零,随机,线性判别分析,主成分分析,SVD奇异值分解等
- Asynchronous API 异步操作
- XML/YAML Persistence XML/YAML/JSON文件存储
- Clustering 分类(kmeans等)
- Utility and system functions and macros 实用程序和系统功能以及宏
- OpenGL interoperability 与OpenGL的交互
- Intel IPP Asynchronous C/C++ Converters 英特尔IPP异步C/C++转换器
- Optimization Algorithms 在有约束和无约束情况下求函数极值
- DirectX interoperability 与DirectX的交互
- Eigen support Eigen(C++线性算数)支持
- OpenCL support OpenCL(面向异构并行)支持
- Intel VA-API/OpenCL (CL-VA) interoperability 英特尔VA-API/OpenCL(CL-VA)的交互
- Hardware Acceleration Layer 硬件加速层
- Parallel Processing 并行处理
- imgproc. Image Processing 图像处理模块
- Image Filtering 图像滤波
- Geometric Image Transformations 几何图像变换(缩放等)
- Miscellaneous Image Transformations 图像变换(二值化等)
- Drawing Functions 绘图
- Color Space Conversions 颜色空间转换
- ColorMaps in OpenCV 颜色映射
- Planar Subdivision 平面细分(OpenCV使用Delaunay算法将平面细分为三角形)
- Histograms 直方图
- Structural Analysis and Shape Descriptors 结构分析和形状描述符
- Motion Analysis and Object Tracking 运动分析与目标跟踪
- Feature Detection 特征检测(霍夫直线霍夫圆等)
- Object Detection 目标检测(模板匹配等)
- Image Segmentation 图像分割(分水岭算法等)
- C API
- Hardware Acceleration Layer 硬件加速层
- imgcodecs. Image file reading and writing 图像文件读写
- C API
- Flags used for image file reading and writing 文件读写参数
- iOS glue 与IOS系统的转换
- MacOS(OSX) glue 与MacOS(OSX) 系统的转换
- videoio. Video I/O 视频流读写
- Flags for video I/O 视频文件读写参数
- Additional flags for video I/O API backends 额外参数
- Hardware-accelerated video decoding and encoding 视频加速编码和解码
- C API for video I/O
- iOS glue for video I/O 与IOS系统的转换
- WinRT glue for video I/O 与WinRT系统的转换
- Query I/O API backends registry 查询I/O API后端注册
- highgui. High-level GUI 界面交互
- Flags related creating and manipulating HighGUI windows and mouse events 交互参数(按钮鼠标事件等)
- OpenGL support OpenGL支持
- Qt New Functions Qt功能
- WinRT support WinRT支持
- C API
- video. Video Analysis 视频分析
- Motion Analysis 运动分析
- Object Tracking 对象跟踪
- C API
- calib3d. Camera Calibration and 3D Reconstruction 摄像机标定与三维重建
- Fisheye camera model 鱼眼相机模型
- C API
- features2d. 2D Features Framework 2D特征框架
- Feature Detection and Description 特征检测和描述(sift,brisk等)
- Descriptor Matchers 描述子匹配
- Drawing Function of Keypoints and Matches 关键点绘制及匹配
- Object Categorization 物体分类
- Hardware Acceleration Layer 硬件加速
- objdetect. Object Detection 目标检测(HOG、QR、Face等haar)
- C API
- dnn. Deep Neural Network module 深度神经网络模型
- Partial List of Implemented Layers 已经实现的网络层(bn,pool等)
- Utilities for New Layers Registration 新的自定义网络层
- ml. Machine Learning 传统机器学习如:Boost、DTrees、KNearest、LogisticRegression、RTrees、SVM等
- flann. Clustering and Search in Multi-Dimensional Spaces 多维空间中的聚类与搜索,包括与FLANN(快速最近邻算法库)的接口
- photo. Computational Photography 计算摄影
- Inpainting 修复算法
- Denoising 去噪
- HDR imaging 高动态范围成像算法,即色调映射、曝光对齐、多曝光相机校准和曝光融合
- Contrast Preserving Decolorization 颜色转换
- Seamless Cloning 无缝克隆
- Non-Photorealistic Rendering 非真实感渲染
- C API
- stitching. Images stitching 在Stitcher类中实现的缝合模块管道
- Features Finding and Images Matching 特征查找和图像匹配ImageFeatures
- Rotation Estimation 旋转估计
- Autocalibration 自动校准
- Images Warping 图像扭曲
- Seam Estimation 焊缝估计
- Exposure Compensation 曝光补偿
- Image Blenders 图像混合器
- gapi. Graph API G-API是OpenCV中的一个特殊模块——与大多数其他主要模块相比,这个模块充当的是一个框架,而不是一些特定的CV算法
- Extra modules:
- alphamat. Alpha Matting 用于从背景图像中提取具有软边界的前景对象
- aruco. ArUco Marker Detection 模块专用于方形基准标记(也称为增强现实标记)。这些标记可用于简单、快速和稳健的摄像机姿势估计
- barcode. Barcode detecting and decoding methods 条形码检测和解码方法
- bgsegm. Improved Background-Foreground Segmentation Methods 改进的背景-前景分割方法
- bioinspired. Biologically inspired vision models and derivated tools 受生物启发的视觉模型和衍生工具
- ccalib. Custom Calibration Pattern for 3D reconstruction 用于三维重建的自定义校准模式
- cudaarithm. Operations on Matrices 矩阵运算
- Core Operations on Matrices
- Per-element Operations
- Matrix Reductions
- Arithm Operations on Matrices
- cudabgsegm. Background Segmentation 背景分割
- cudacodec. Video Encoding/Decoding 视频编码解码
- cudafeatures2d. Feature Detection and Description 特征检测和描述
- cudafilters. Image Filtering 图像过滤
- cudaimgproc. Image Processing 图像处理
- Color space processing
- Histogram Calculation
- Hough Transform
- Feature Detection
- cudalegacy. Legacy support
- NPPST Core
- NPPST Image Processing
- NPPST Signal Processing
- cudaobjdetect. Object Detection 目标检测CascadeClassifier、HOG
- cudaoptflow. Optical Flow 光流
- cudastereo. Stereo Correspondence 立体通信
- cudawarping. Image Warping 图像扭曲
- cudev. Device layer 硬件层
- cvv. GUI for Interactive Visual Debugging of Computer Vision Programs 用于计算机视觉程序交互式可视化调试的GUI
- datasets. Framework for working with different datasets 用于处理不同数据集的框架
- Action Recognition
- Face Recognition
- Gesture Recognition
- Human Pose Estimation
- Image Registration
- Image Segmentation
- Multiview Stereo Matching
- Object Recognition
- Pedestrian Detection
- SLAM
- Super Resolution
- Text Recognition
- Tracking
- dnn_objdetect. DNN used for object detection 用于目标检测的DNN
- dnn_superres. DNN used for super resolution 用于超分辨率的DNN
- dpm. Deformable Part-based Models 基于可变形零件的模型
- face. Face Analysis 人脸分析
- freetype. Drawing UTF-8 strings with freetype/harfbuzz 使用freetype/harfbuzz绘制UTF-8字符串
- fuzzy. Image processing based on fuzzy mathematics 基于模糊数学的图像处理
- Math with F0-transform support
- Math with F1-transform support
- Fuzzy image processing
- hdf. Hierarchical Data Format I/O routines 分层数据格式I/O例程
- Hierarchical Data Format version 5
- hfs. Hierarchical Feature Selection for Efficient Image Segmentation 用于高效图像分割的分层特征选择
- img_hash. The module brings implementations of different image hashing algorithms
The module brings implementations of different image hashing algorithms.
该模块提供了不同图像哈希算法的实现 - intensity_transform. The module brings implementations of intensity transformation algorithms to adjust image contrast 该模块提供了用于调整图像对比度的强度变换算法的实现
- julia. Julia bindings for OpenCV OpenCV的Julia(一种新的编程语言)绑定
- line_descriptor. Binary descriptors for lines extracted from an image 用于从图像中提取线条的二进制描述符,这个模块提取的是一种新的信息,不同于点信息,具有尺度不变性
- mcc. Macbeth Chart module 颜色图表模块
- Color Correction Model 颜色校准模块
- optflow. Optical Flow Algorithms 光流算法
- ovis. OGRE 3D Visualiser 3D可视化,ovis是ogre3d的简化渲染包装器。在API中使用了Ogre术语,并且假定Ogre脚本用于高级定制
- phase_unwrapping. Phase Unwrapping API 相位展开API,相位图
- plot. Plot function for Mat data Mat数据的点位
- quality. Image Quality Analysis (IQA) API 图像质量分析
- rapid. silhouette based 3D object tracking 基于轮廓的三维目标跟踪
- reg. Image Registration 图像配准,使用像素来计算配准,而不是特征值
- rgbd. RGB-Depth Processing RGBD图像处理
- saliency. Saliency API 显著性API,寻找图像的显著区域加快检测
- sfm. Structure From Motion 从运动中构造,从2D到3D的多视图重建
- Conditioning
- Fundamental
- Input/Output
- Numeric
- Projection
- Robust Estimation
- Triangulation
- Reconstruction
- Simple Pipeline
- shape. Shape Distance and Matching 形状距离与匹配
- stereo. Stereo Correspondance Algorithms 立体匹配算法(块匹配)
- structured_light. Structured Light API 结构光API
- superres. Super Resolution 超分辨率
- surface_matching. Surface Matching 曲面匹配,3D ppf算法 icp算法
- text. Scene Text Detection and Recognition 自然场景图像中的文本检测和识别
- tracking. Tracking API 跟踪API,一个接口,可以插入多种算法并对它们进行比较
- videostab. Video Stabilization 视频稳定模块
- Global Motion Estimation 运动估计
- Fast Marching Method 快速匹配
- viz. 3D Visualizer 3D窗口可视化
- wechat_qrcode. WeChat QR code detector for detecting and parsing QR code 微信解码
- xfeatures2d. Extra 2D Features Framework 额外的2D匹配框架
- Experimental 2D Features Algorithms
- Non-free 2D Features Algorithms
- Experimental 2D Features Matching Algorithm
- ximgproc. Extended Image Processing 图像处理进阶
- Structured forests for fast edge detection
- EdgeBoxes
- Filters
- Superpixels
- Image segmentation
- Fast line detector
- EdgeDrawing
- Fourier descriptors
- Binary morphology on run-length encoded image
- xobjdetect. Extended object detection 目标检测进阶WaldBoost detector
- xphoto. Additional photo processing algorithms 其他照片处理算法,修复,白平衡等