Abstract: According to the characteristics of infrared dim targets in strong clutter background, this paper presents a method of detecting frared dim signals based on adaptive filtering, which denoises the images first, then eliminates the background based on adaptive filtering enhanced target signals and deletes the false targets based on the determination principles of point target imaging signals characteristics from testing. The algorithm effectively solves the contradiction of the photoelectric detection equipment between high detection probability and low false alarm rate. The test results show that this method can effectively extract the dim signal targets with the small area SNR 4 from the single frame images, and that the detection probability is not less than 0.75, and the false alarm rate is not heigher than 1 time at 100 frames.