Comparative analysis of Tesseract and Google Cloud Vision for Thai vehicle registration certificate

Optical character recognition (OCR) is a technology to digitize paper-based document digital form. This research studies the extraction of characters from Thai vehicle registration certificate via Google Cloud Vision API and Tesseract OCR. The performance both OCR APIs also examined. 84 color image files comprised three sizes/resolutions five characteristics. For suitable type comparison, greyscale binary are converted images. Furthermore, pre-processing techniques, sharpening, contrast adjustment, brightness applied enhance quality before applying two APIs. was evaluated in terms accuracy readability. results showed that works well for with an 84.43%, whereas 47.02%. highest came 1024×768 px, 300dpi, using sharpening adjustment as techniques. In readability, has more readability than Tesseract. proposed conditions facilitate possibility implementation system.