University System of Maryland

Robert Vinson

Yang Tao

Gary E Seibel

Dongyi Wang

Maxwell Holmes

Xuemei Cheng

An automated method for crab meat picking which has an image processing system configured with an imaging sub-system and an image processing sub-system acting in cooperation with each other where the image processing system operates in accordance with a processing routine. An image of a crab to be processed is obtained by the imaging sub-system which is transmitted to the image processing system where the processing routine determines appropriate cut lines to be made on the crab being processed based on instructions from the image processing sub-system. The crab being processed is then cut in accordance with the appropriate cutting lines developed.

Automated crab meat picking method

Intelligence guided system and method for fruits and vegetables processing includes a conveyor for carrying produces, various image acquiring and processing hardware and software, water and air jets for cutting and controlling the position and orientation of the produces, and a networking hardware and software, operating in synchronism in an efficient manner to attain speed and accuracy of the produce cutting and high yield and low waste produces processing. The 2nd generation strawberry decalyxing system (AVID2) uniquely utilizes a convolutional neural network (AVIDnet) supporting a discrimination network decision, specifically, on whether a strawberry is to be cut or rejected, and computing a multi-point cutline curvature to be cut along by rapid robotic cutting tool.

Automatic vision guided intelligent fruits and vegetables processing system and method

A vision guided intelligent system for automated crab meat picking operates in a fully automated or semi-automatic modes of operation using a crab meat picking routine based on (a) the CNN model localization of back-fin knuckles algorithm, and/or (b) the Deep Learning model which accurately locates not only knuckle positions, but also crab legs and crab cores, with a high pixel accuracy (up to 0.9843), and low computation time. The subject system uses a concept of analyzing crab morphologies obtained from digital crab images, and, using a Deep Learning architecture integrated in the system, segments crab images into five regions of interest in a single step with high accuracy and efficiency. The image segmentation results are used for generating crab cut lines in XYZ and angular directions, determining starting cutting points in Z plane, and guiding cutting tools and end effectors to automatically cut crabs and harvest crab meat.

Growing community of inventors

Rockville, MD, United States of America

Robert Vinson