Yingjie Xia

A method of automatic recognition for answer sheet

The traditional method of recognizing an answer sheet is to use optical mark reader (OMR). A kind of OMR only recognizes a certain answer sheet with fixed format, which results in the poor universality of OMR. We propose a recognition method for answer sheet with arbitrary format. After designing the new answer sheet or using the existing ones, the printed answer sheets will become images by high-definition (HD) scanning after being filled in an exam. And the images of answer sheets will be recognized automatically by image processing techniques. According to the positioning cross found in answer sheets, the images will be corrected if they are tilted. Then candidate number recognition, option recognition and page number recognition will be carried out in the order specified by users. The method of maximum between-cluster variance will be used for candidate number recognition and option recognition. On the other hand, the page number of answer sheet will be recognized by template matching. Experimental results show that the accuracy can reach 100%. And this method can be realized easily, the cost is low, and it has good universality.

High Conductivity of Mg-Doped Al0.3Ga0.7N with Al0.4Ga0.6N/AlN Superlattice Structure

The highly conductance of Mg-doped Al0.3Ga0.7N layer using low-pressure metal organic chemical vapour deposition (MOCVD) on Al0.4Ga0.6N/AlN superlattice structure was reported. The rapid thermal annealing (RTA) has been employed for the effective activation and generation of holes, and a minimum p-type resistivity of 3 Ω·cm for p-type Al0.3Ga0.7N was achieved. The RTA annealing impacted on electrical, doping profile and morphological properties of Mg-doped Al0.3Ga0.7N with Al0.4Ga0.6N/AlN superlattice structure have been also discussed. The quality of Mg-doped Al0.3Ga0.7N with Al0.4Ga0.6N/AlN superlattice structure degraded after annealing from HRXRD. At appropriate annealing temperature and time, surface morphology of Mg-doped Al0.3Ga0.7N can be improved. A step-like distribution of [Mg] and [H] in p-type Al0.3Ga0.7N was observed, and thermal diffusion direction of [Mg] and [H] was also discussed.

Assembly method for multimodal image registration

Assembly method for automatic multimodal image registration is put forward. There are three methods: majority domination, weighted domination and concentration criterion. The first one stems from the simple principle: the minority shall be subordinate to the majority. The second one comes from the fact that different algorithms in general have different rates of correctness. We call these two dominations as weighted assembly registration criterion (WARC) or weighted majority criterion (WMC). The third one may be helpful in choosing a seemingly reasonable registration result by seeking the maximum points in a back projection image when the above two criteria do not work well. By introducing the techniques of edge detection and image equalization, six different registration methods based on normalized mutual information (NMI) and maximization of alignment metric (AM) are developed for multimodal image registration between infrared thermal images and corresponding visible light images. Experiments show the adoption of these criteria may achieve the better registration than any proposed individual method.

A Method of Feature Matching Based on Multi-Pose Face

In this paper, SIFT algorithm is combined with AdaBoost algorithm, and a method of feature matching based on multi-pose face is put forward. Firstly, the face region is extracted from multi-pose face images by AdaBoost. Secondly, SIFT characteristic vectors of the main regions are matched. The images of the ORL face DB are used in this paper, and some pictures taken in the experiment are used too. The matching results are acceptable and reasonable. It can be used to research face recognition based on multi-pose face.

A synthetic registration method for multi-modal images based on dynamic credibility

A synthetic registration method based on dynamic credibility is proposed to investigate infrared images and their visible light counterparts. Mutual information and alignment metric are adopted together with edge detection and equalization to form a synthetic method composed of six individual metric. Dynamic credibility is put forward to measure the correctness and the essentiality of the six methods during the course of integration. It is proved that the synthetic registration method based on dynamic credibility is more accurate than any individual registration method. And the synthetic registration method has high robustness. Besides, the credibility can be corrected with the ceaseless registrations for multi-modal images. It has strong adaptability.