Design of an Intelligent Fruit Recognition Weighing System based on Deep Learning

With the development of society, people's requirements for food quality and safety are constantly improving, and fruit, as an important component of daily consumption, its weighing method needs to be upgraded intelligently. Traditional weighing of fruits is mostly a manual operation, which has the problems of low recognition efficiency and unable to count the weight automatically. This paper proposes an intelligent fruit identification and weighing system design based on OpenCV. The system takes Raspberry Pi 4B as the core, combines the image acquisition with CSI camera, classifies and identifies the fruit types through the CNN algorithm in OpenCV, and obtains the weight of the fruit using HX711 gravity sensor to achieve automatic identification, weight counting and billing functions. The system can also upload the total price of fruits to WeChat applet, which is convenient for managers to view in real time and later inventory. Experimental results show that the intelligent fruit identification and weighing system has an identification accuracy of more than 90 %, with automatic weighing, price output and other functions, which can effectively improve work efficiency and reduce human error, and has good application and promotion value.