Detect and identify multiple objects in an image by cropping each object separately and querying Google Lens for individual identification.
MIT
650,521 tools. Updated 2026-10-14 00:42
"OpenCV" matching MCP tools:
- Verifies Cloudflare Turnstile challenges by clicking the checkbox using OpenCV template matching. Supports custom template images for non-English versions.MIT
- Auto-finds and clicks Cloudflare Turnstile checkboxes using a three-tier detection strategy: primary selectors, secondary container search, and optional OpenCV template fallback.MIT
- Probes an RTSP stream or local file to retrieve width, height, and fps, and generates a candidate test case list. Optionally accepts per-frame annotations to include discovered labels.MIT
- Extract text from image files (PNG, JPG, TIFF, BMP) using multiple OCR engines and preprocessing options.MIT
- Check which Python third-party libraries are installed on the device before importing them in scripts. Lists packages available in the AScript App environment.MIT
Matching MCP Servers
- AlicenseNot gradedqualityFmaintenanceProvides OpenCV computer vision capabilities for AI assistants, enabling image processing, object detection, face detection, and video analysis through natural language.15 PyPI111MIT
- AlicenseAqualityBmaintenanceProvides image and video analysis capabilities for LLMs, with local preprocessing (ffmpeg/OpenCV) and any OpenAI-compatible vision model for understanding and Q&A.4MIT
- Find matching visual regions in a retained screenshot by comparing a crop from another screenshot, returning candidate positions with similarity scores.MIT
- Solve Cloudflare 'Verify you are human' challenges by detecting and clicking the checkbox widget. Use this when a page is stuck on a browser-checking interstitial.MIT
- Aligns a photo of a hand-drawn sketch to its map layout, extracts the selected ink color, and appends georeferenced polygons or polylines to an existing feature class.Apache 2.0
- Locate visual elements on a page using image template matching when DOM selectors are unavailable. Returns coordinates of best match.MIT
- Analyze a trading card image using vision and centering analysis to return PSA/Beckett subgrades, overall condition score, and ROI verdict.Business Source 1.1
- Detect anti-bot walls on the current page and automatically apply the best bypass, including Cloudflare Turnstile verification, returning JSON with detection and bypass results.MIT
- Execute Python code to perform any task: HTTP requests, data processing, file operations, system commands, image manipulation, and more with auto-install of needed packages.MIT
- Capture screenshots of the active browser tab with support for full-page stitching, region clipping, and format selection. Options include base64 encoding for vision models and automatic resizing to avoid dimension limits.MIT
- Automatically install Python dependencies for HWP document editing. Choose full install with OCR or lightweight core-only mode.-
- Extract chemical structures from images and return SMILES with confidence scores and bounding boxes. Optionally detects text labels via OCR.MIT
- Reads a web page and generates two thematically matching images (landscape and portrait) without text or recognizable faces, using a local image generation model.-
- Converts native or scanned Korean PDFs into editable HWP/HWPX documents, reconstructing tables, images, and layout for visual similarity.-
- Erase unwanted objects from video frames spatially: SAM2 segments the target from point prompts, OpenCV Telea fills the region per frame, with silent output.-
- Refine a PNG mask by applying erosion and Gaussian blur, reducing edge artifacts and smoothing boundaries for cleaner video compositing.-
- Find where a small pattern image (a button, icon, logo or crop) appears inside a larger screenshot or picture, using OpenCV cv2.matchTemplate. Returns the best position, plus every non-overlapping position scoring above the threshold, as pixel boxes with scores. Use it for exact-looking pixel patterns at the same scale; for different scale, rotation or perspective use /v1/homography. Price: $0.003 a call (3 free calls a day without an API key).ConnectorNo auth