基于神经网络的不均匀光照下围棋识别方法

Weiqi Recognition Method under Uneven Illumination based on Neural Networks

针对围棋机器人系统中光照不均匀造成的棋子识别误差问题,提出了基于神经网络的不均匀光照下围棋棋子识别方法。首先,利用边缘检测算法确定棋盘位置;其次,利用阈值分割法检测出棋盘四个标定点的坐标,并通过透视变换矩阵转换为标准棋盘图像;然后,基于多层感知器搭建神经网络模型,并训练出分类器;最后,将标准棋盘图像利用分类器进行棋子分类,并输出识别结果。所提方法在强光、室内灯光、暗光三种光照不均匀情况下对围棋棋子的平均识别准确率可达到98%以上,实验结果验证了该方法的有效性。

In view of chess pieces recognition error caused by uneven illumination in Weiqi robot system,this paper proposes a Weiqi recognition method based on neural networks.Firstly,edge detection algorithm is used to determine the checkerboard position;secondly,the coordinates of the four calibration points of the checkerboard are detected by threshold segmentation method and transformed into a standard checkerboard image by perspective transformation matrix.Then,the neural network model is built based on multilayer perceptron and the classifier is trained.Finally,chess pieces in the standard checkerboard image is classified by the classifier and the recognition result is output.The average recognition accuracy of the proposed method can reach more than 98%under three kinds of uneven illumination conditions,such as strong light,indoor light and dark light.The experimental results have verified the effectiveness of the proposed method.