Research On Intelligent Detection Technology Of Ship Water Gauge Based On Image

With the continuous deepening of human exploration of the ocean,the shipping industry has become increasingly developed.Since the weight of coal,cement,steel and other cargoes transported by ships cannot be easily measured by ordinary weighing instruments,the water gauge based on Archimedes’ principle has become a very critical link in measuring the value of commodities.The traditional water gauge weighing method is mainly realized by professional observers observing the position of the waterline on the six sides of the ship.Due to the lack of data backtracking methods,this method is very susceptible to human subjective factors,wind and waves,light and other objective factors.At the same time,with the development of international trade,the increase in the value of goods also brings certain requirements to the accuracy and stability of the water gauge weighing method.This paper mainly studies the intelligent application of image processing technology in the field of ship water gauge detection,and uses the method based on image processing and detection to intelligently detect ship water gauge.The hardware system is mainly composed of a water gauge image acquisition module composed of industrial cameras,industrial lenses and pan/tilt,a data transmission module composed of wireless transmission networks and handheld terminals,and an image processing module composed of computers.The software system mainly includes the water gauge detection algorithm part and the user graphical interface part.This paper conducts a comprehensive study on a variety of water gauge detection algorithms,including traditional algorithms such as image enhancement,image segmentation,template matching,feature extraction,etc.,and proposes an image-based water gauge intelligent recognition algorithm,which has a certain generality and stability Sex.First,the collected ship water gauge video is sampled every other frame,image correction,image enhancement and other pre-processing work are carried out,and the grayscale image is converted into a grayscale image to draw a grayscale histogram.The OTSU threshold segmentation algorithm is used to complete the binarization of the water gauge image,which is convenient Improve recognition accuracy.Finally,a combination of morphological opening and closing operations is used to remove most of the salt and pepper noise while filling the holes of the defective characters.The horizontal and vertical bidirectional histogram projection of the binary image of the water gauge realizes the positioning and segmentation of characters.At the same time,after studying the method of recognizing characters based on the template matching method based on grid pixel distribution characteristics,a secondary character recognition algorithm is proposed.SVM classifier is used to train the recognition model based on HOG feature extraction.The recognition characters are roughly classified and then re-recognized by the secondary check algorithm based on optimized SURF feature matching.The recognition accuracy is much higher than that of the traditional algorithm.In the process of waterline detection,research is carried out from the perspective of edge detection and color features.In view of the various shortcomings reflected by the above-mentioned waterline recognition algorithm in actual use,this paper applies the clustering segmentation algorithm to the recognition of ship waterline.A method based on the K-means clustering segmentation algorithm to identify the waterline is proposed,which can accurately segment the water surface and the hull,and significantly improve the segmentation accuracy.Finally,statistical analysis is performed on the results obtained under different frame numbers to eliminate outliers and reduce the recognition error.After realizing the intelligent detection software of water gauge,a large number of field acquired water gauge image data were tested,and the accuracy of the results was analyzed from different angles.Based on the overall design of draft detection,this paper studied the draft character detection algorithm and the waterline recognition algorithm deeply,and built a relatively perfect intelligent draft detection system on the basis of this.The detection experiments on the ship water gauge images collected on the wharf show that the intelligent ship detection technology studied in this paper can effectively realize the intelligent recognition of the water gauge,and the recognition results are in line with expectations.