The recognition accuracy of the traditional visual image analysis method for the end face of the steel bar depends on the intensity of illumination and the level of image pixels.However,the changeable and uneven illumination of the site environment will greatly reduce the recognition accuracy of the steel bar contour.The method of deep learning is adopted to establish a reliable im-age training model,so that the detection system has the ability to learn the steel bar contour image independently,and constantly opti-mize the training parameters,so as to achieve the effect of improving the recognition accuracy of the steel bar end face contour.
关键词
深度学习/神经网络/激活函数/训练模型/自主学习/Kmeans算法
Key words
deep learning/neural network/activation function/training model/independent learning/Kmeans algorithm
维普
万方数据