Abstract: Nowadays, the development of information technology is keeping in a fast speed. Large amount of information are exchanged too active to identify the original owner. A serious problem of digital contents copyright violation is happening today and the most effective manner to avoid it is digital watermark technology. In this work, a new approach for digital watermark embedding is presented. This method is different from the traditional ways which are pixel-exchanging and frequency-domain-based watermarking scheme. It combines the pixel-exchanging scheme and frequency-domain-based scheme by exchanges the lower bit plane and imitates the imaging process of a true-lens-imaging system respectively. On the other hand, This method use the point spread function as secret key that keeps the security and spread the embedded information to all bit plane of the image that keeps the robustness. Numerical simulation examples show the effect.
Abstract: Nowadays, digital imaging device has become a popular item of our daily life. Even a cell phone can gives a high density image. People can enjoy taking a photo at anytime and anywhere, but such a mechanically simplified imaging device usually cannot mount an optical lens filter. Therefore the digital filtering effect is needed. Many image effect filters have been released by image processing soft maker or digital camera manufacture, but most of them were made just for funny, the resulting image is unnatural. This research focuses on the digital filtering effect, and it is realized according to a real optical lens imaging system. In this work, we prepared several optical filter lenses, and take photos with or without using them. Assuming the photo without using a filter is the real object, and then the photo using the filter is the observed filtering image, the digital filter can be derived by a division theoretically. Using these digital filters, the resulting filtering images are gotten. From these result, we find that some of the filtering effect is good, but some of the effects were not enough. Therefore, assuming the cause of these problems from the experimental results, show the direction for future research such as improving experiment approach or adopting nonlinear algorithms is also attempt.
Abstract: Science of dealing with the prevention and treatment of disease is important for the maintenance of health in human social life. Laser technology plays an important part in various fields of medical science. In general, a spot laser light is used in laser therapy because of the limitation of laser output. Therefore, the operator has to move the spot position by hand when the treatment area is larger then the laser light spot. We propose a method to convert a laser light to match the target treatment surface using time-divided computer-generated hologram (CGH). In order to get a high laser light efficiency, we consider the phase-only type CGH. We use discrete reconstruction CGH to overcome the speckle problem in the generation CGH. A spatial continuous laser light distribution can be obtained by superimposing the discrete reconstruction from several time-divided CGHs.
Abstract: Remote sensing of position detection and motion detection plays an important part in the welfare filed of human social life. Privacy is a serious problem when detecting human position and motion. A position sensitive device (PSD) only detects the position of a light target without detailed image of human. For detecting multiple light targets, there are various problems such as synchronized light source and noise in daily life space. In this study, we suggest a method using modulated light targets to separate different light targets and eliminate the noise in daily life space for motion detection of a PSD. The target light intensity is modulated with different high frequency source for different targets, so that the noise in daily life space can be eliminated and the positions of different targets can be separated by using band-pass filters. Experimental results showed this method is effective.
Abstract: We propose a wearable supporting system with a CMOS image sensor for the visually impaired people in operating capacitive touchscreen. This system attaches the CMOS image sensor without a lens to the tip of the middle finger. The icons and buttons displayed on the touchscreen are replaced to the color barcodes. Touching the surface of the touchscreen with the CMOS image sensor directly, the color barcode is detected and decoded. The decoded results are returned to the user by some interaction like audio. Then, the user touches the button area around the color barcode by the forefinger to operate the target device. This system can provide very easy and natural way for operating the touchscreen to the visually impaired people who usually recognize the materials by the finger. Any mechanical modification of the target device is not needed. The modification can be made by changing its software program. Since the color barcode is sensed by the image sensor without any lens touching the surface of the touchscreen, each bar in the color barcode should be blurred. So, we develop an easy and simple image processing to handle such problem. We design it as the hardware module to achieve the high performance and low-power wearable device. A prototype hardware using an FPGA shows the hardware size, the performance and the actual demonstration.
Abstract: People living in the information age, are more and more attention to their lives. It is also said, social life is more important in present and future. The social life contains three fields. In this paper, we propose a new model for active contours to detect objects in a given medical image, in order to facilitate people to have medical treatment. The proposed method is based on techniques of piecewise constant and piecewise smooths Chan-Vese Model, semi-implicit additive operator splitting (AOS) scheme for image segmentation. Different from traditional models, our model uses the level set which are corresponding to ordinary differential equation (ODE). Our model has more improved characteristics than traditional models, such as: less sensibility of noise; unnecessary of re-initialization and high speed by the simplified ordinary differential function. Finally, we validate the proposed model by numerical synthetic and real images. The experimental results demonstrate that our model is at least two times more efficient than the widely used methods.
Abstract: People living in the information age, are more and more attention to their own lives. It is also said, social life is more important in present and future. The social life contains three fields. In this paper, we will propose a new method for adjunctive therapy in social life. Recent years, as the bandelets transform has some benefits, many scholars are interested in this field. They proposed many methods to solve different problems in different fields. In this paper, we propose a new maximum local energy method to calculate the low coefficients of images. And then adopt the sum modified laplacian method to select the high coefficients of images. Later, we compare the results with wedgelets transform. In our experiments, we take wedgelets transform, bandelets transform, and LE-wedgelets transform for comparing the results. Beside the human vision, we also compare the results by quantitative analysis. The numerical experiments state clearly that the maximum local energy is an effect way for image fusion, which can get well performance in visual effect and quantitative analysis. During 100 clinic CT/MR fusion experiments in practice, compare with previous methods, the PSNR of our method is improved respectively 5.836, 5.337, 0.035.
Abstract: Digital X/γ-ray imaging technology has been widely used to help people deliver effective and reliable security in airports, train stations, and public buildings. Nowadays, luggage inspection system with digital radiographic/computed tomography (DR/CT) represents a most advanced nondestructive inspection technology in aviation system, which is capable of automatically discerning interesting regions in the luggage objects with CT subsystem. In this paper, we propose a new model for active contours to detect luggage objects in the system, in order to facilitate people to identify the things in luggage. The proposed method is based on techniques of piecewise constant and piecewise smooths Chan-Vese Model, semi-implicit additive operator splitting (AOS) scheme for image segmentation. Different from traditional models, the fast implicit level set scheme (FILS) is ordinary differential equation (ODE). Characterized by no need of any pre-information of topology of images and efficient segmentation of images with complex topology, the FILS scheme is fast more than traditional level set scheme 30 times. At the same time, it performs well in image segmentation of DR images in our experiments.
Abstract: Recently, various user interfaces are developed. However the operation of user interface is very difficult for the physically handicapped persons who cannot move their hand. The stationary user interface we are proposing uses head tracking via a camera and a display. It is portable and can operate household appliances. It is also operated intuitively in head tracking.
Abstract: Texture surface analysis is very important for machine vision system. We explore Gray Level Co-occurrence Matrix-based 2nd order statistical features to understand image texture surface. We employed several features on our ground-truth dataset to understand its nature; and later employed it in a building dataset. Based on our experimental results, we can conclude that these image features can be useful for texture analysis and related fields.