OpenCV Overview

Relevant source files

• 3rdparty/fastcv/fastcv.cmake
• 3rdparty/ippicv/CMakeLists.txt
• CMakeLists.txt
• cmake/OpenCVCRTLinkage.cmake
• cmake/OpenCVCompilerOptimizations.cmake
• cmake/OpenCVCompilerOptions.cmake
• cmake/OpenCVFindLibsGUI.cmake
• cmake/OpenCVFindLibsPerf.cmake
• cmake/OpenCVFindLibsVideo.cmake
• cmake/OpenCVModule.cmake
• cmake/OpenCVPCHSupport.cmake
• cmake/OpenCVUtils.cmake
• cmake/templates/cvconfig.h.in
• modules/core/CMakeLists.txt
• modules/core/include/opencv2/core.hpp
• modules/core/include/opencv2/core/base.hpp
• modules/core/include/opencv2/core/core.hpp
• modules/core/include/opencv2/core/cv_cpu_dispatch.h
• modules/core/include/opencv2/core/cv_cpu_helper.h
• modules/core/include/opencv2/core/cvdef.h
• modules/core/include/opencv2/core/matx.hpp
• modules/core/include/opencv2/core/operations.hpp
• modules/core/include/opencv2/core/private.hpp
• modules/core/include/opencv2/core/types.hpp
• modules/core/include/opencv2/core/utility.hpp
• modules/core/include/opencv2/core/utils/buffer_area.private.hpp
• modules/core/src/algorithm.cpp
• modules/core/src/buffer_area.cpp
• modules/core/src/command_line_parser.cpp
• modules/core/src/precomp.hpp
• modules/core/src/system.cpp
• modules/core/test/test_operations.cpp
• modules/core/test/test_utils.cpp
• modules/highgui/CMakeLists.txt
• modules/java/CMakeLists.txt
• platforms/linux/riscv64-071-gcc.toolchain.cmake
• samples/android/CMakeLists.txt
• samples/android/color-blob-detection/CMakeLists.txt

OpenCV (Open Source Computer Vision Library) is a high-performance C++ library for real-time computer vision and image processing modules/core/include/opencv2/core.hpp62-67. It provides a modular framework for handling image data, mathematical computations, and hardware-accelerated inference.

For a detailed exploration of the source code organization, see Repository Tour and Module Map.

Repository Layout and Versioning

The repository is organized into a core library, optional functional modules, and platform-specific build scripts. Version management is handled via CMake and propagated to core headers, with version detection from Git metadata if available CMakeLists.txt123-132.

Directory	Contents
modules/	Functional modules (core, imgproc, dnn, etc.) modules/core/CMakeLists.txt41.
cmake/	Build infrastructure, dependency detection, and module macros cmake/OpenCVModule.cmake1.
platforms/	Platform-specific configurations (Android, iOS, etc.) CMakeLists.txt94-99.
3rdparty/	Bundled dependencies (zlib, libjpeg, libpng, etc.) CMakeLists.txt148-164.
apps/	Developer tools like annotation, interactive calibration, and visualization utilities.

For a detailed tour of the top-level layout, directory contents, and the OpenCV 5.x module split (geometry/calib/stereo/ptcloud/features), see Repository Tour and Module Map.

Sources: CMakeLists.txt1-132, modules/core/CMakeLists.txt1-41, cmake/OpenCVModule.cmake1-25

Module Architecture and Dependencies

OpenCV is structured as a collection of interdependent modules. Every module depends on opencv_core, which defines the fundamental Mat and UMat structures modules/core/include/opencv2/core.hpp64-66.

Module Dependency Graph

The following diagram illustrates canonical dependencies between major modules and how they bridge from low-level data structures to high-level vision tasks.

Code Entity Mapping:

• opencv_core: Defined in modules/core/CMakeLists.txt modules/core/CMakeLists.txt41.
• opencv_imgproc: Image processing foundations.
• opencv_highgui: High-level GUI abstractions modules/highgui/CMakeLists.txt1.
• ocv_add_module: CMake macro in cmake/OpenCVModule.cmake cmake/OpenCVModule.cmake124.

Sources: modules/highgui/CMakeLists.txt3-7, cmake/OpenCVModule.cmake124-127, modules/core/CMakeLists.txt41-43

Module Declaration via ocv_add_module

Modules are registered using the ocv_add_module macro cmake/OpenCVModule.cmake124. This macro handles dependency resolution, categorization, and binding generation:

• Required Dependencies: Modules that must be present for this module to build cmake/OpenCVModule.cmake86.
• Optional Dependencies: Modules that enhance functionality if present cmake/OpenCVModule.cmake88.
• Wrappers: Specifies which language bindings (Python, Java, JS, ObjC) should wrap this module cmake/OpenCVModule.cmake93.

Sources: cmake/OpenCVModule.cmake77-117, cmake/OpenCVModule.cmake124-155

Build System Architecture

OpenCV uses a sophisticated CMake-based build system that manages cross-platform compilation, hardware optimization, and third-party library integration CMakeLists.txt1-166.

Configuration Flow

The configuration process is split into two passes:

1. Initial Pass: Collects module information and descriptions cmake/OpenCVModule.cmake130-155.
2. Target Creation: Resolves dependencies and creates the actual CMake targets cmake/OpenCVModule.cmake172-223.

System Initialization Diagram:

Sources: CMakeLists.txt42-52, [CMakeLists.txt:139], cmake/OpenCVModule.cmake54-74, cmake/OpenCVUtils.cmake1-15

For instructions on configuring, building, selecting toolchains, and running sanity checks, see Getting Started: Configure, Build, and Sanity-Check.

Platform and Backend Support

OpenCV abstracts platform differences through backends, particularly for GUI, video input/output, and parallel execution.

GUI Backends (highgui)

The highgui module selects a windowing backend based on target platform and available libraries modules/highgui/CMakeLists.txt43-192:

• Wayland / GTK: Linux graphical backends modules/highgui/CMakeLists.txt55, modules/highgui/CMakeLists.txt165.
• Win32 UI: Native Windows UI modules/highgui/CMakeLists.txt152.
• Cocoa: macOS native UI modules/highgui/CMakeLists.txt145.
• Qt: Cross-platform backend supporting Qt4, Qt5, and Qt6 modules/highgui/CMakeLists.txt80-99.

Performance Libraries

OpenCV integrates several third-party libraries for acceleration and parallel processing:

• TBB: Intel Threading Building Blocks for parallel loops modules/core/CMakeLists.txt51-53.
• IPP: Intel Integrated Performance Primitives for low-level optimizations modules/core/CMakeLists.txt64-81.
• CUDA: NVIDIA GPU acceleration support via opencv_cudev and CUDA modules.
• Eigen: Linear algebra optimization backends.

Sources: modules/highgui/CMakeLists.txt43-192, modules/core/CMakeLists.txt51-81

Documentation, Samples, and Developer Apps

OpenCV maintains comprehensive API documentation, extensive code samples, and developer utilities:

• Documentation: Generated via Doxygen from annotated C++ headers.
• Samples: Located in samples/ and categorized by language and task domains.
• Developer Apps: Include tools for interactive camera calibration, pattern generation, and diagnostics.

For details on how documentation is produced, where tutorials and samples live, and developer-facing applications, see Documentation, Samples, and Developer Apps.

Sources: CMakeLists.txt1-10

Navigation Guide

This wiki is structured to mirror the codebase's directory and module topology:

1. Build System (Section 2): CMake configuration, module registration (ocv_add_module), and package generation.
2. Core Module (Section 3): Fundamental data structures (Mat, UMat), SIMD universal intrinsics, and CPU dispatch.
3. Functional Modules (Sections 4-10): imgproc, dnn, features2d, videoio, highgui, and geometric/3D modules.
4. Language Bindings (Section 11): Code generation pipeline for Python, Java, JavaScript, and Objective-C/Swift.
5. Testing & Infrastructure (Sections 12-18): Google Test integration (ts), GPU acceleration, and domain-specific tools.