Treffer: Object segmentation and its visual quality assessment for images

Ph.D. ; In this era of information explosion, more and more information is presented in a form of images. Hence, analysis and processing of image content become very important. There is no doubt that the objects in an image play a primary role in understanding the image content. In order to extract the objects, object segmentation is a key technique. Therefore, it is necessary to develop better object segmentation algorithms for generating accurate segmentation results. Moreover, since object segmentation is a pre-processing step, different applications have their own requirements for it. However, current commonly used statistics-based object segmentation quality assessment methods are not able to satisfy all applications. Therefore, speci c quality assessment methods should be designed for the intended application. Although the object segmentation technique has evolved over the past decades, there is still much room for improvement and the quality assessment method is less studied. The main objective of this thesis is to investigate quality assessment methods in terms of human visual perception and object segmentation algorithms. Two main parts of this work are briefly summarized below. ; In the first part, we explore visual quality assessment of object segmentation in terms of subjective evaluation and objective measure. Firstly, we present a subjective object segmentation visual quality database, in which a total of 255 segmentation results were evaluated by more than thirty human subjects. This database is used to evaluate the performance of the objective measures and analyze their pros and cons. Then, we propose a supervised objective measure for an object segmentation visual quality evaluation, which involves four human visual properties. Finally, our measure is compared with some state-of-the-art objective measures on our database. The experiment demonstrates that the proposed measure performs better in matching subjective judgments. ; The second part concerns interactive object segmentation. According to ...