Underwater Object Tracking By Image Enhancement And Feature Fusion

Given the complex underwater optical imaging process and low image quality, it is difficult to extract the features of the underwater object to be tracked accurately. Meanwhile, the difficulty of underwater object tracking is further exacerbated by the high degree of freedom in the motion of the underwater object, which is often characterized by morphological changes and rapid movements. To address the above problems, the paper proposes an underwater object tracking method with image enhancement and feature fusion(IEFF). A novel underwater image enhancement method based on color correction and light attenuation prior is added to efficient convolution operators(ECO) for tracking. The object feature extraction and sample space model are also improved in ECO. Experimental results on the self-built database show that the area under the curve of success ratio for our method is better than ECO and other methods.