Algorithm 2 Mean-shift Based Candidate Target Region Identification

Input: The input image $f$ and the binary result of the iterative CFAR approach $g$.

Process: (1)Initialize the selection matrix $p$ as $g$. Set the size of the selected region $L\times L$, the radius of the searching region $r$, the side length of the candidate target region $L$ and the maximum width of the ship to be detected $W$. (2)Take out the potential ship pixels whose values are one in $g$, and sort them according to their intensities in descending order. (3)Take the potential ship pixels in sequence. For each potential ship pixel, check $p$ to see if the pixel can be taken as a selected point. When it can be, select it as a start point and move to the next step, otherwise repeat step (3) to take the next potential ship pixel. (4)Do the mean-shift operation until convergence. (5)For the final selected point, Check $p$ to see if it can be taken as a selected point. When it can be, go to step (6), otherwise go to step (3) to take the next potential ship pixel. (6)The $L\times L$ region centering on the final selected point is detected as a candidate target region. Employ the $l_1$ norm regression to extract the principal axis of the target, and identify the valid points of the target. Update the selection matrix $p$ by setting the corresponding pixels of valid points in $p$ as zero, so that the valid points of the target are no longer taken as either the start or the final selected point. (7)Go to step (3) to take the next potential ship pixel.

Output: Candidate target regions.