OpenCV Computer Vision Skill
Comprehensive computer vision and image processing using OpenCV WebAssembly.
Trigger When
使用此技能的场景:
- 图像处理 / Image Process:resize(调整大小)、blur(模糊)、threshold(阈值)、morphology(形态学)、sharpen(锐化)、crop(裁剪)、stats(统计)
- 目标检测 / Detection:cascade(级联检测)、dnn/YOLO(深度学习检测)、hog、contour(轮廓)、circle(圆检测)、line(直线检测)
- 特征提取 / Features:orb、sift、akaze、match(特征匹配)、homography(单应性)、optflow(光流)、template(模板匹配)
- 图像拼接 / Stitching:panorama(全景拼接)、stitch(多图拼接,扫描件模式)
- 照片处理 / Photo:denoise(降噪)、inpaint(修复)、stylize(风格化)、hdr、pencil(素描)、detail(细节增强)
- 相机标定 / Calibration:chessboard(棋盘格)、undistort(去畸变)、corners(亚像素角点)、pose(位姿估计)
- 机器学习 / ML:train-knn(训练KNN)、train-svm(训练SVM)、predict(预测)、kmeans(聚类)
- 批处理 / Batch:batch-process(批量处理)、batch-convert(格式转换)、batch-stats(批量统计),支持多线程
Prerequisites
- Required plugin:
openclaw-wasm-sandbox - WASM file:
~/.openclaw/skills/opencv/files/opencv-component.wasm- Download if missing:
openclaw wasm-sandbox download \ "https://raw.githubusercontent.com/guyoung/wasm-sandbox-openclaw-skills/main/opencv/files/opencv-component.wasm" \ "~/.openclaw/skills/opencv/files/opencv-component.wasm"
- Download if missing:
Tool
Use wasm-sandbox-run tool with the WASM file:
wasm-sandbox-run({
wasmFile: "~/.openclaw/skills/opencv/files/opencv-component.wasm",
args: ["<operation>", ...args]
})
For file access, use workDir or mapDir to expose directories to the sandbox.
Operations Overview
| Category | Operations | |----------|-----------| | Image Process | resize, blur, threshold, morphology, sharpen, crop, stats | | Detection | cascade, dnn (YOLO), hog, contour, circle, line | | Features | orb, sift, akaze, match, homography, optflow, template | | Stitching | panorama, stitch | | Photo | denoise, inpaint, stylize, hdr, pencil, detail | | Calibration | chessboard, undistort, corners, pose | | ML | train-knn, train-svm, predict, kmeans | | Batch | batch-process, batch-convert, batch-stats |
Detailed Usage
See references/operations.md for complete operation reference with arguments, examples, and sandbox notes.
Scan to join WeChat group