A soil sample box number identification system and method based on an artificial neural network

The invention relates to the field of geotechnical test equipment, and provides a soil sample box number recognition system and method based on an artificial neural network, and the system comprises an image collection device, an image segmentation device, a character segmentation device, and a character recognition device. The image acquisition device is used for acquiring images in an image acquisition area of the sample conveying device; The image segmentation device carries out character region segmentation on the collected image and extracts a region where the box number is located; The character segmentation device carries out preprocessing and character segmentation on an area where the extracted box number is located; The character recognition device establishes a character template database through self-training and recognizes characters through feature comparison. According to the invention, the artificial neural network is used for replacing human eyes to observe and identify the sample box number, the network can be automatically trained in the identification process, a better identification result is obtained, the reading error generated during manual observation is avoided, the box number reading accuracy is ensured, and the mechanization and automation of the device are improved.