Font Size: a A A

Research And Implementation Of The Container Image Recognition And Locating System

Posted on:2014-01-04Degree:MasterType:Thesis
Country:ChinaCandidate:D ChenFull Text:PDF
GTID:2248330398474527Subject:Software engineering
Abstract/Summary:PDF Full Text Request
As the rapid development of economy and foreign trade, In shipping transport, containers are used more frequently for its advantages. Container scale will be larger and Container will be used more efficient. At the same time, container crane grows higher and higher, so that it’s harder to locate the container for the crane drivers, and it will cost more time to locate the container. The worst case, it will cause to accidents to make huge losses. Traditional manual control mode blocks the economic development.Container auto-landing technology is still in exploration. For auto-landing system, we expect that the containers’position can be located automatically using machine vision under the different environment, and the system can give the optimized routing which directs the spreader to move. As the result, we need to move forward deeply to research on the container auto-landing system using image recognition technology.Image recognition has been widely used in the fields of aerospace, medical, communications, industrial automation, robotics and military. Based on image recognition technology, I propose a method to recognize the container with SVM. Firstly, preprocess the container image, and separate the different image segmentation. Secondly, Extract the feature vectors of container image includes geometry, color, texture vectors. Finally, recognize the container from the image using SVM method. Container image recognition technology is useful in container detection and location, and it is also very important for improving the operating efficiency of the port.For the sake of getting the container’s position, first, extract the plate top of the container. Second, calculate the transformation parameters including translation, rotation, scale between the plat top of the container as a reference plane and the spreader plat as movement plane. As the result, the transformation parameters are the optimized routing to direct the spreader to move. It’s significant that container automatic positioning technology is used to auto-landing system to improve handling efficiency and reduce labor intensity.
Keywords/Search Tags:Auto-landing System, Image Recognition, Texture Extraction, HSV, GeometryFeature, Position Estimation
PDF Full Text Request
Related items