Research On Application Of Ship Rust Identification And Rust Removal Path Planning

At present,the derusting operation of large ships in China is mainly carried out by workers with hand-held nozzles spraying dry sand.The manual sand blasting derusting method has the disadvantages of heavy pollution,high cost,low efficiency and dangerous operation.The environment-friendly and efficient ship rust removal equipment and technology are monopolized by European and American developed countries.The domestic ship rust removal equipment technology is not mature,and the research focuses on the rust removal equipment.There is little research on ship rust point identification and rust removal planning,which is far from the actual engineering application.Due to the limitation of environmental protection policy and labor cost,there is an urgent need for environmental protection and efficient rust removal intelligent equipment which can replace manual rust removal.In this paper,the scheme of using abrasive water jet to remove rust is put forward.The specific research is as follows.Firstly,according to the actual working environment of ship rust removal,the overall technical scheme of the whole set of rust removal equipment is designed,and the function is designed and the rust removal scheme is planned according to the actual demand.Secondly,the image acquisition in the dock is affected by light.MSRCR algorithm is used to preprocess the image based on Retinex theory,and the color comparison of images is enhanced by balancing illumination.The ship corrosion is divided into large area corrosion and point corrosion.The manual threshold segmentation method and the maximum between-class variance method are used to comprehensively process the image to complete the identification of ship rust.Referring to the concept of topological map and grid map,the concept of rust point map is proposed innovatively.The rust point map is constructed based on the rust point recognition image.The rust point center coordinates are transformed from pixel coordinate system to map coordinate system by scale transformation.The rust point area is replaced by the minimum rectangular area containing rust traces,and the rust point map containing rust point area and rust point center coordinates is constructed.Then,the derusting path planning is carried out based on the rust map,and the derusting path planning problem is transformed into a TSP problem and the problem is modeled.Then,aiming at the unknown number of rust points in the local area in the actual work,an improved genetic algorithm is proposed.The nearest neighbor algorithm is used to construct the initial population with the same number of individuals and rust points,and the evolutionary reversal and translational transformation operators are added to increase the population diversity and the stability of the algorithm.The algorithm results are stable and can significantly shorten the convergence rate.Subsequently,the experiment proves that the improved genetic algorithm can effectively plan the path of rust map.Finally,the rust points of simulated ship corrosion area and actual ship corrosion image are identified,and the rust point map is constructed.Then the improved genetic algorithm is used for rust removal path planning.The results show that the algorithm can effectively identify and process the rust points in different situations,and make reasonable planning.