Embedded System Based Raspberry Pi 4 for Text Detection and Recognition

Detecting and recognizing text from natural scene images presents a challenge because the image quality depends on conditions in which is captured, such as viewing angles, blurring, sensor noise, etc. However, this paper, prototype for detection recognition proposed. This based Raspberry Pi 4 Universal Serial Bus (USB) camera embedded our model, was developed using Python language. Our model deep learning detector through Efficient Accurate Scene Text Detector (EAST) localization Tesseract-OCR, used an Optical Character Recognition (OCR) engine recognition. controlled by Virtual Network Computing (VNC) tool computer via wireless connection. The experiment results show that rate captured can reach 99.75% with low computational complexity. Furthermore, more performant than Tesseract software terms of rate. Besides, it provides same performance huge decrease execution time average 89% compared to EasyOCR board.