机器视觉系统超分辨率图像准确识别方法研究
Research on the method of super-resolution image recognition in machine vision system
-
摘要: 为了能够准确识别超分辨率图像, 提出了一种基于图像重构的神经网络图像识别方法.这种超分辨率图像识别方法采用Gamma方法去除机器视觉系统超分辨率图像中的无用信息, 根据超分辨率图像的阈值对图像进行分割, 并提取超分辨率图像的特征, 重构分割后的超分辨率图像, 利用Hopfield神经网络实现对机器视觉系统超分辨率图像的识别.仿真实验结果证明, 所提方法能够对机器视觉系统超分辨率图像进行准确识别, 并且识别效率高、速度快.Abstract: In order to accurately identify super-resolution images, a neural network image recognition method based on image reconstruction is proposed. This super-resolution image recognition method uses Gamma method to remove the useless information in the super-resolution image of the machine vision system. The image is segmented according to the threshold of the super-resolution image, and the features of the super-resolution image are extracted, and the segmented image is reconstructed. Super-resolution image, using the Hopfield neural network to realize the super-resolution image recognition of machine vision system. The simulation experiment results show that the proposed method can accurately recognize the super-resolution image of the machine vision system, and the recognition efficiency is high and the speed is high.