Zhengzhi Wang

Single-image shadow detection and removal using local colour constancy computation

This study is concerned with the problem of shadow detection and removal from single images of natural scenes. In this work, the authors propose a shadow detection method with a surface descriptor, termed colour-shade, which allows them to include the physical considerations derived from the image formation model capturing gradual colour surface variations. The authors incorporate a colour-shade descriptor into the condition random field model to find same illumination pairs and to obtain coherent shadow regions. The authors propose a shadow removal method using an improved local colour constancy computation, which uses anisotropic diffusion to estimate the illuminant locally for each image pixel in shadow. The authors evaluate their method on two shadow detection databases. The experimental results demonstrate that their shadow detection and removal method is state of the art.

Semi-automatic object segmentation using colour invariance and Graph cuts

Conventional semi-automatic or interactive methods, which require a small amount of user inputs for region segmentation of objects, have obtained the best segmentation results. A new semi-automatic segmentation technique by using coloured scale-invariant feature transform (CSIFT) to extract seed pixels in Graph Cuts is introduced here. First, CSIFT is used to extract feature points of objects in the image. Then, a voting process is used to extract the matched points as object seeds. The detailed technique via s–t Graph Cuts has been presented, and a new segmentation energy cost function with two colour-invariant descriptors has been proposed: colour-name descriptor and colour-shade descriptor. The colour-name descriptor introduces high-level considerations resembling top-down intervention, and the colour-shade descriptor allows us to include physical consideration derived from the image formation model capturing gradual colour surface variations and provides congruencies in the presence of shadows and highlights in the segmentation. The experimental results prove that the proposed method provides high-quality segmentations with object details.

Reinforcement Learning for Image Understanding

Reinforcement Learning is one of the hottest issues in current AI research fields. Its a effective method in solving some machine learning problems. Its high efficiency, simpler programming, easier understanding, and better performance. Here I will share my understanding. If there are something wrong, thanks for correct. In reinforcement learning, the learner is a decision-making agent that takes actions in an environment and receives reward (or penalty) for its actions in trying to solve a problem. After a set of trial-and-error runs, it should learn the best policy, which is the sequence of actions that maximize the total reward.

Subpixel Edge Detection: An Improved Zernike Orthogonal Moments Method

In this paper, an improved method for detecting edge points is presented. We expatiated on the principle of Zernike moments and the method of sub pixel edge detection Based on Zernike moments. With deducing of the rotated Zernike moments, a new way to acquire the perpendicular distance from the center of the circular to the edge (l) is introduced in the criterion for detection of the edge points. Experimental results show that the improved sub pixel edge detection algorithm can detect the edge effectively and precisely. It has higher precision, integrality of detection and stronger robustness to noise. Besides, it also provides a valuable and promising reference for practical use of edge detection techniques.

Automatic color image segmentation using CSIFT and Graph Cuts

In this paper, we propose a new automatic color image segmentation method using Colored Sift (CSIFT) and Graph Cuts. Color provides valuable information in object segmentation and recognition tasks. However, color information is vulnerable to be affected by shadows and highlights. CSIFT is a stable and distinctive feature with respect to variations in the photometrical imaging conditions. It has been demonstrated that the CSIFT is more robust than the conventional SIFT with respect to color and photometrical variations. On the other hand, Graph Cuts is proposed as a segmentation method of a detailed object region. But it is necessary to give seeds manually. In our method, the object is recognized first by CSIFT interest points. After that, the object region is cut out by Graph Cuts using CSIFT interest points as seeds.