Applied Mechanics and Materials Vols. 263-266

Abstract: This paper Study problems which based on lifting wavelet transform image processing. Coding and decoding a complete digital image by using W97-2 wavelet basis wavelet transform, combined with the embedded zerotree wavelet coding and binary arithmetic coding, and complete a lossless compression combined with the international standard test images. Experimental results show that graphics, image processing will come into a higher level because of wavelet analysis combined with image processing.

Abstract: The commonly used JPEG and GIF images on Internet web sites have become more popular for steganographic applications. However, the use of lossy image compression will destroy the embedded message on the marked images. This article presents a novel data hiding method for palette images which can embed double messages, secret and robust messages, into palette images. The receivers can extract both the embedded two messages with no loss if the stego-image is not encountered any attack. If the stego-image is attacked, we can still extract the robust message with little distortion from the attacked images. The proposed method is based on parity check and statistical quantity histogram. Experiment results show the proposed method can achieve high performance in embedding capacity and robustness.

Abstract: This paper which is based on digital image process technology, with LabVIEW and IMAQ Vision,designs a effective method about image acquisition and processing. The system achieve image acquisition,save,playback,gray transformation,denoise,threshold segmentation ,template matching and calculation the center of the target lacation.Virtual Instrument which own flexible hardware and software,has greatly enhanced efficiency and accuracy.

Abstract: In this paper, we propose a high capacity reversible data hiding method for error diffused halftone images. It is a challenge to conceal data in halftone images due to their smaller depth. We utilize statistical features of pixel block patterns to embed data. The marked images can be perfectly reconstructed into their original halftone images. Traditional pair-based schemes selected two similar patterns as one pair to toggle one-bit data embedding. We extend the approaches and present a group-based encoding scheme to collect similar patterns into groups for multi-bits embedding. Experimental results show the proposed method can significantly improve the capacity of marked halftone images while preserve the similar marked image quality.

Abstract: In order to solve the problem of the photo response non-uniformity (PRNU) for High Dynamic Range (HDR) video system, this paper discusses the causes of the problem and the common solution, and then proposed a new non-uniformity correction approach based on the reference source for HDR video system. This approach use high-pass filter to calibration image in Frequency domain to get the correction matrix of one level light conditions at first. And then performs an inverse Fourier transform to the spatial domain. Finally calculate the difference between the correction matrix and the target image to obtain the corrected results. The experimental result indicates that the approach has good effect on dealing with non-uniformity of the HDR images and the definition has been greatly improved.

Abstract: Nowadays, medical image processing and three-dimensional visualization have become a very important support for doctor diagnosis and treatment planning. It’s a novel technology that using ITK and VTK to process and display medical images in VC++. In this paper, the Curvature FlowImageFilter class of ITK libraries is used to denoise and smooth the medical images. The MC algorithms is used to reconstruct the volume data in the VS2008. Immediate Mode algorithms and Stripper filter algorithms are adopted to speed up large data processing. The experiment results demonstrate that using the MC algorithms and the acceleration algorithms base on the VTK can implement 3-D visualization efficiently and satisfy practical applications.

Abstract: This paper presents a quadrilate extraction method based on four vertices of the image object to solve the problrm of object area inaccuracy depending on conventional rectangular segmentation method. Firstly, four vertices are given by the mouse-click depending on quadilateral object spatial feature; and then the quadrilateral object shape feature is estimated by each side of the quadriangle computing by straight-line formulation of two adjacent vertices of object in order to acquare the quadrialeratal object inernal pixel. At last, the pixels of quadrialeratel external region are cleaned. Arbitrary quadrilateral region segmentation is achieved accurately. The method can be used for any quadrilateral region segmentation based on image processing, The experiments on actual color parking images show any parking cell template extraction accuracy reached 100% and parking cell state detection accuracy is 99.8%. The result verifies the algorithm effectiveness and robustnes

Abstract: Edge-detection is the basic characteristic of images. Edge-detection plays an important role in computer vision and image analysis, it is the key link of image analysis and recognition. In this paper, we mainly analyze several edge-detection operators, research the processing results. The edge-detection operators mainly include Roberts operator, Prewitt operator, Sobel operator, Canny operator and LoG operator. Finally we discuss the advantage and disadvantage of the edge-detection operators.

Abstract: In this literature, an automatic white balance adjustment algorithm based on Gaussian decomposition is proposed. The improved algorithm is aiming at handling the problem occurred in automatic white balance algorithm, such as lack of correction or high complexity. First, Gaussian distribution of gray level is obtained by using Gaussian decomposition. And then, gain curve of each component in RGB color image are calculated by the parameters. After the coefficient is used for white balance adjustment, a corrected image is finally obtained. Experimental results show that the algorithm can effectively correct the image color cast problems, and are suitable for a variety of different scenarios.

Abstract: The edge information of pupil was extracted by the least-square method, and the iris outer circle was extracted by the improved Canny Operator plus Hough Transform. The segmental-secondary linear localization method adopting edge detection and Radon Transform was proposed to remove the noise from eyelid on the eyelid localization, the eyelash noise and eyelid shadows were removed by threshold method.