Research On Vision And Control System Of Chess Robot

With the rapid development of industrial automation in the world and the continuous improvement of science and technology level,robot technology has gradually become a frontier technology and research hotspot.It's the latest research in the science,materials,pattern recognition,robotics,intelligent control,mechanics,bionics,and so on,and it's an important symbol of the age of human intelligence.As an entertainment robot,chess robot can independently observe and recognize chess games like a human being,think about playing strategies independently,and complete moves by arms.This is based on the technology of image processing,depth learning,and the design and implementation of the chess robot vision and control system.Firstly,a method of segmentation based on color characteristics and an improved binary image filtering method is proposed.According to the color characteristics of the characters of the chess pieces,the preliminary segmentation was carried out.The image denoising after segmentation was carried out by using the binary image filtering method.Linear fusion and expansion calculation were carried out.It use that camera model of pinhole and zhang zhengyou to calculate the world coordinate of chess pieces and get the coordinate of robot;Secondly,a recognition algorithm of chess pieces based on digital coding is proposed,which can recognize chess pieces without image processing and reduce the system running time;In addition,a chess recognition method based on convolutional neural network is proposed.Based on the AlexNet network structure,the network structure suitable for chess recognition is adjusted,and a large number of chess data sets are used to train the network to achieve high accuracy of chess recognition.This method can be applied to the chesses with different fonts.It does not need to set the characteristics of the chessmen manually,which improves the recognition efficiency and universality of the chesses.Finally,the effectiveness of the proposed method is verified by experiments with relevant hardware devices.