A Method for Infrared Sea-Sky Condition Judgment and Search System: Robust Target Detection via PLS and CEDoG

It has always been a serious challenge to efficiently detect infrared targets from remote sea-sky background without any prior knowledge. This is especially true when target features of different areas are irrelevant in the same image. The main contribution this paper design an algorithm judge condition and target, which two-direction local maximum (TLM) method peak singularity (PLS) proposed, constant excitatory difference Gaussians (CEDoG) operator applied by analyzing characteristics target. First foremost, TLM adopted refine seasky area extract suspected line, strategy “de-false”to realize accurate determination line. Secondly, small (pixels 2 × 9 9) detected applying PLS method. Finally, CEDoG filtering used suppress improve significance high threshold OTSU (H-OTSU) find most significant self-designed growth rule ensure accuracy integrity maritime detection. Comparing with other state-of-the-art methods experiments, our robust effective performance terms recall, precision, elapsed time, complexity, detection false alarm rate.