Advanced Materials Research Vols. 760-762

Abstract: Adaptive infrared image contrast enhancement is presented based on modified particle swarm optimization (PSO) and incomplete Beta Function. On the basis of traditional PSO, modified PSO integrates into the theory of Multi-Particle Swarm and evolution theory algorithm. By using separate search space optimal solution of multiple particles, the global search ability is improved. And in the iteration procedures, timely adjustment of acceleration coefficients is convenient for PSO to find the global optimal solution in the later iteration. Through infrared image simulation, experimental results show that the modified PSO is better than the standard PSO in computing speed and convergence.

Abstract: According to the coordinates of the particled graph, the paper presents a secondary pretreatment method which is invariant to translation, scaling and rotation. Having the advantages of efficient and accurate, the new method is significant to image processing. As the core of the pretreatment, orientation pretreatment is described in detail. In combination with the semi-structured data storage model, the new pretreatment method can help to achieve image content retrieval and data mining of a deeper level.

Abstract: Palm vein pattern recognition is one of the newest biometric techniques researched today.At present, literatures selecte the center of the palm as the ROI of palm vein recognition. However the vein image in this area is not clear in some peoples palm. In this paper, we proposed a new location method of ROI which takes thenar area as the ROI. In the experiment part, it compares the recognition rate between the new and the traditional ROI in self-established contactless palm vein database. The result shows that this new method has got the recognition rate of 98.9258% and has increased recognition rate 2.0911% compared with the traditional one.

Abstract: This paper puts forward an improved image motion compensation method according to the complex mechanical system and expensive cost problem when the pendular aeronautical camera uses multiple sensors to compensate image motion. This method can only use a linear position sensor to obtain a linear output voltage signal and a sinusoidal output voltage signal, thus a linear position sensor can not only complete image motion compensation function but also complete pendulum sweep range positioning. Finally the experiments has proved that the method this paper has chosen not only simplifies the mechanical system design and reduces the aerial camera cost, but also can meet the accuracy and real-time requirements of various pendular aeronautical camera or space camera.

Abstract: Palm vein pattern recognition is one of the newest biometric techniques researched today. In order to reduce the recognition calculation and increase the recognition accuracy, the region of interest (ROI) in the palm vein image must be selected. There are three areas on the palm: the center area of palm, the thenar area and the hypothenar area.This paper does contrast experiment of light absorption on the three area of palm. The experiment result is the thenar area is selected as the main scope of ROI.

Abstract: Palm vein is covered with skin, some captured NIR palm vein images are not clear because of light scattering in the skin. The identification performance is not high by extraction palm vein feature directly. This paper proposes project the palm vein image matrix based on two dimensional Fisher linear discriminant directly, then calculates the Euclidean distance of the projection matrix, seeks the nearest distance for classification. The experiment has been done in a self-build palm vein database. Firstly, this paper tests the proposed method by different component numbers, then compares the identification performance with the other typical palm vein recognition methods.

Abstract: mage enhancement has applied widely in biomedical, nondestructive testing, satellite remote sensing and other fields. Especially for the low contrast radiographic images, usually there are some disadvantages for a radiographic image such as the local area image does not show a striking contrast. In order to improve the clearness of low contrast radiographic images, in this paper we combined global adaptive equalization with local dynamic enhancement,then we simulate this enhancement algorithm. The new method will not only effectively increase the global contrast of low contrast radiographic images, but also intensify local details. Because the new algorithms contrast enhancement coefficient function can be adjusted dynamically and locally, the new algorithms are not only adaptive to the process of radiographic images but also having great reference value to the other grayscale images.

Abstract: A new method for quick identification of images of mature potatoes under the soil background is reported in this thesis. Firstly, the segmentation problem in multidimensional space is transformed into one dimensional space by transforming the color image to gray image. Then, the optimum threshold is obtained on the base of the Otsu threshold segmental method. Lastly, the image is identified by the improved algorithm, which is based on the ABS algorithm. The merits of this algorithm are that it is simple with highly efficient identification. Experimental results show that the method can avoid interference of clods completely and identify potatoes from the soil background effectively.

Abstract: Matrix-Transpose is one of the most basic matrix operation. The Corner Turning Matrix (CTM) is a very important part of the real-time image system. In the radar real-time image system, transpose is needed to do in the distance compression and the azimuth compression. The speeds, reading and writing, of the traditional transpose algorithm arent match, for reading operation with low efficiency. In this paper, a Block CTM algorithm based on DDR3 SDRAM is dedicated and used in the real-time image system. Through the contrast between the traditional transpose algorithm and the Block CTM algorithm, the proposed method improves the efficiency of the radar real-time image system greatly.

Abstract: Staff line removal is a key step before segmentation and recognition of music image and plays an important role in OMR (Optical Music Recognition) research, the result of staff line removal directly influences the performance and function of the whole OMR system. However, over-removal and under-removal often occurs in the processing and leads to the low efficiency of music recognition rate. So, in order to solve the arduous problem, an approach based on run-length graph slice and topological structure of music is put forward by careful analysis of staff line and music notation structure. Experience results show the validity and practicality of the presented algorithm fast and effectively.