Research On Detection Technology For Infrared Image Target Based On Adaptive Morphological Filters

Infrared Target Recognition and Measure is the core technology of accurate guiding weapon system. However, Infrared Target works in the expansion with long-distance, only being small points and some pixels in screen, and its lum is even lower than bright clouds in the background. There exists some difficulty to be solved, such as low signal noise ratio, no information about texture, little dimension and figure.This paper is made up of two parts: One is about method of infrared image processing; The other is about the Digital Signal Processing in real time.In image processing software: To the complex background of image, Infrared Target is easy to lose, so morphological filters for the target detection and recognition is proposed in this paper, and explain the morphological filter's principle and property in particular. But only one structure element can't fit with the various background, so this paper Genetic Algorithm to optimize the multi-structure element of morphological filters to improve the target detection and recognition ability. And then, it analyzed the affection of operator for Genetic Algorithm in advantage. Finally, this paper explore self-adaptive genetic algorithm with morphology filter. The target detection is intelligent, which can self-adaptation adjust structure element to fit environment alternation. The result shows that this method can detect small Infrared Target in complicated background more effectively.In Infrared Target Measure hardware: To the complicated calculator, a real-time high resolution image processor is triumphantly developed. The memory is large and the precision is high, the processor has capability of parallel and pipeline processing and the operational speed is 1600MIPS. Based on the hardware platform, the processor was used for infrared target detecting.