Design of Image Fuzzy Recognition Algorithm Based on FPGA

Object recognition is an important component of machine vision, and it is a worthwhile research direction to let machines distinguish natural objects like human beings. In view of the real-time and accuracy of item recognition in machine vision, this paper proposes an algorithm design of image fuzzy recognition based on FPGA. The system adopts the PANGO platform Logos series PGL22G development board, equipped with OV5640 camera as the image data input. Through white balance calibration and single-pixel blocking algorithm, it can detect the maturity of common fruits and count the number of mature fruits. Experimental results show that under the condition of processing 30 frames of 640×480 resolution images, the system can identify mature fruits in the same category of fruits and obtain the total number of mature fruits. In 1000 pictures tests, the comprehensive recognition accuracy of objects is 85.2%, and the accuracy of quantity recognition is 99%. It has a wide range of application scenarios in agricultural fields such as orchards.