The Image Fusion Algorithm Based On Wavelet Transform And PCNN

With the rapid development of computer vision and image sensor technology, image fusion technology in the corresponding field got great development and application. More focus on image fusion is an important branch of computer vision applications. Optical lens focusing range is quite limited in use process; it also promoted the rapid development of the technology. The technology can effectively improve the accuracy of target detection, and subsequent image recognition and edge detection, image segmentation, feature extraction and so on. In target identification, microscopic imaging, military operations, machine vision, and other fields, more focus on image fusion has been widely used.In this paper, main work is as follows:1. To focus on research and development of image fusion, more must be reviewed, understand and know the multi-source image fusion is the all levels of the basic principle and method, and the fusion effect the objective evaluation index of performance are introduced.2. Fully understand the image fusion method based on wavelet transform, and through the corresponding experiment was validated.3. To understand the pulse coupled neural network (PCNN) model, and by using the modified pulse coupled neural network applied in medical image fusion, the paper proposed an effective block medical image fusion method based on adaptive pulse coupled neural networks (PCNN). Source images are divided into several blocks, and then we calculate the spatial frequency (SF) of the blocks as linking strength β of the PCNN, so it adjusts β of the PCNN adaptively. The block images are input into PCNN to get the oscillation frequency graph (OFG), which expresses the quality of the block images, so we can fuse the clear part of the source images. The experimental results show that the block medical image fusion algorithm is more efficient than other common image fusion algorithms, and prove the adaptive PCNN method is effectively as well.4. Finally, this paper presents a new method for color image fusion. According to human visual characteristic, perception brightness and objective brightness is a logarithmic relations, so if based on the similar area brightness through PCNN fusion, can effectively improve the image gradient, enhanced image edges, can significantly improve the image level. Specific implementation is the-RGB image transformation to the HIS color model, I component is executed wavelet decomposition, the corresponding component of the decomposed is input into PCNN model, by fusing new I component is obtained. Also to the corresponding H and S components according to some fusion rules to fusion, and then by inverse transformation to RGB space, the fused image is obtained. The experimental results show that this proposed method is of the convergence performance and the objective evaluation index is better than other traditional methods.