Researching Of Detection Algorithm Of Sea Infrared Ship Target Based On Mathematical Morphology

The effective detection for small targets in low SNR image is always a hot research field. In order to improve the SNR, amend the image quality, suppress imaging interfere and imaging, the IR image pretreatment is a very important process. It decides that if the hardware system can exactly detect and track the target at any time. Aimed at this problem, It is mainly studies that Mathematical morphology is applied in the background suppression and the edge detection for small targets IR image in this paper. Focus on the application of mathematical morphology in the infrared images of the background suppression and edge detection, and improved mathematical morphology on the basis of the results of previous studies.The mathematical morphology preprocessing algorithm performed with the improved morphological filters for the original image. A frame element bigger than the target and a smaller one are used by turns in the opening operation; search results from two different filter images as a result of pretreatment. Experiments showed that the effect of filter morphology in both the treatment and processing time together has an advantage over other algorithms. Then, on the basis of improving the preprocessing results of the morphological, a new method that improving the multi-structure complex morphological edge detection algorithm be used, in order to further reduce the loss of edge information.the subtract output is the background suppression result, It is proved that the improved morphological filter is better than others in the effect and managed time. Then the improved morphological filter is used in the background suppression result. Use the improved morphological erode edge extraction in order to reduce the edge information that erode extraction dismissed. The experiment results indicate that the improved method can resist noise better, distill the edge of target exactly, reserve more detail information of the target. the edge is detected more sustainable than the traditional morphological methods, The edge of detection is smoother than traditional morphological edge detection. After detection of pre-processing the edge of the original image, all kinds of noise and aberrance affect to the target area of image eliminate furthest. The efficiency and veracity to the targets is improved.