Image segmentation and feature extraction are the premise for machine vision system to analyze and identify the image. Threshold image segmentation algorithm according to the method of two dimension threshold has a lot of calculation in calculating the threshold, and the minimum error threshold method can not use the spatial information of image. This paper presents an improved quantum-behaved particle swarm optimization based on the night segmentation and feature extraction technology. This paper introduces the QPSO algorithm based on multi group and multi stage improvement. The QPSO optimizing algorithm gradually approaches the global optimum threshold value to achieve better convergence and stability. An algorithm of vision image segmentation and feature extraction based on improved quantum-behaved particle swarm optimization is designed. Experimental results show that the optimization process of this algorithm has less control parameters and faster convergence speed.

This paper describes rough neural network which consists of a combination of rough neurons and conventional neurons. Rough neurons use pairs of upper and lower bounds as values for input and output. In some practical situations, it is preferable to develop prediction models that use ranges as values for input and/or output variables. Integrating rough set theory with neural network theory, a novel information fusion method based on rough neural network is proposed to fuse the different-source images in agricultural robot. It is used to fuse infrared and visible images in order to take full advantage of the complementary information between infrared and visible images. Experimental results show that the fusion effect and speed are both better than standard wavelet transform and the conventional neural network.