Image Processing Techniques For Automatic Detection System In The Ship's Draft

Water gauge is used for weighing goods of low price, difficult in ponderation, or large quantity in volume. This method has scientific principle and veracity to some extent, and has been approbated by the world. Its result can be regarded as bases of handing over and taking over and balancing of goods, processing counterclaim, counting carriage, customs declaration.The paper primarily does the research on how to get the position of the waterline inside the scale number relative to the bottom edge. The result produced by computers need to be combined with human measuring. That is to say, by human recognizing the scale number and computer estimating the position, we can increase the accuracy greatly.Because the used images are quite large in size, for the purpose of reducing running time, the paper firstly takes the binary segmentation in the hue space, then gets the round position of the boundary of the ship board and the water, and cut the image based on this processing. Secondly, the image is cut again by searching the character of the water gauge's scales above the water line.This paper takes two steps to detect the waterline edge: 1) the edge detection approach based on the direction region distance measure(DRDM) for color images using in the HSI color space; 2) the heuristic edge pick-up approach based on the grads extent are used.For waterline edge selection, this paper adopts a voting method to choose the possibly true edges, in which the value of the following attribute are large: the hue grads, the saturation grads, the intensity grads, and color DRDM grads and the times of searching edges. According to the difference of the average values between the upside area and the downside, the method select the true edge of waterline. Finally, the horizontal line is fitted by the least square method.After the analysis of the characters of the scale number o the shipboard, this paper presents a new segmentation method , binary segmentation line-by-line, and obtain the better result of the segmentation. Then we get positions of objects by projection. At last, according to the rule of the water gauge, we reason out the expression to calculates the position of the waterline inside the scale number relative to the bottom edge by using information mentioned above. After the multiple experiments on many images, it is proved that this method is effective.