I need a small, self-contained program that takes an image of a printed betting ticket, recognises its contents, and instantly tells me whether that ticket is a winner or a loser. The tool must then append a new line to a single CSV file capturing the ticket’s: • winning status (win / lose) • kind of bet (single, combo, etc.) • odds shown on the slip • market (e.g. 1X2, BTTS, over/under) • name of the match or event If you can also pick up the ticket number or stake while you are already parsing the slip, feel free to include those columns, but they are optional. Input will always be an image—either dragged in, captured from a webcam, or selected from disk. Because of that, I expect you to rely on a solid OCR or image-recognition stack (Tesseract, OpenCV, EasyOCR or a similar library) so the software works with standard printed slips under normal shop lighting. Once the data is parsed, the program should automatically cross-check the event result (an open API or another lightweight method you prefer is fine) and decide the outcome of the bet. No manual typing should be necessary after the image is supplied. Acceptance criteria: 1. Accuracy: OCR must correctly read the required fields on at least 95 % of clear, unobstructed tickets. 2. Speed: total processing time—image import to CSV write—under 5 seconds on a mid-range Windows PC. 3. Portability: deliver a single-click desktop executable; no command-line or complex setup for shop staff. 4. Data integrity: the CSV file grows safely with each scan and never overwrites existing records. Send back the finished executable, source code, and a brief README explaining any third-party APIs or libraries used so I can maintain the system in-house.