Object Recognition Using a Dynamic Learning Approach

Most of the traditional object recognition systems consist of the training phase and the testing phase. Once the modules have been built at the training phase, these modules can't be adjusted anymore. So if there're other new images to be added later, it is necessary back to the training phase to retraining a new module. In addition, these systems aren't designed for mobile devices that are difficult to move around and inflexible. - In view of this, this book proposes a system with SEG as the front-end equipment and a back-end object recognition system to identify an object. This book is divided into two parts: moving object segmentation and object recognition. For the first part, in order to improve the accuracy of object recognition, this book integrates optical flow, CamShift, and GrabCut to achieve a high accuracy by shaking the object to be recognized in hand. In the second part, in the dynamic learning process, simultaneously based on the quality of recognition result to determine which strategy to adopt to adjust the database module to the best state at any time. In addition, the system also uses the function of Google search by image, to recognize those untrained images.

Autorentext

Zhuang Bi-Yao, Master of Science: Majored in Computer Science & Engineering at National Sun Yat-Sen University. Serve as founder and CEO of LuckyCSE, and work as Image processing R & D personnel at Wistron Corp., Kaohsiung.