With the popularization of high-definition (HD) images and videos (1920×1080 pixels and above), there are even 4K (3840×2160) television signals and 8 K (8192×4320) ultrahigh-definition videos. The demand for HD images and videos is increasing continuously, along with the increasing data volume. The storage and transmission cannot be properly solved only by virtue of the expansion capacity of hard disks and the update and improvement of transmission devices. Based on the full use of the coding standard high-efficiency video coding (HEVC), super-resolution reconstruction technology, and the correlation between the intra- and the interprediction, we first put forward a “division-compensation”-based strategy to further improve the compression performance of a single image and frame I. Then, by making use of the above thought and HEVC encoder and decoder, a video compression coding frame is designed. HEVC is used inside the frame. Last, with the super-resolution reconstruction technology, the reconstructed video quality is further improved. The experiment shows that by the proposed compression method for a single image (frame I) and video sequence here, the performance is superior to that of HEVC in a low bit rate environment.
Video-frame registration is a critical step in the video super-resolution process. However, many existing registration methods can handle only small local neighborhood movements or overall affine transformations. This limits the reconstruction of fast moving or significantly deforming objects. We propose a super-resolution registration method for video reconstruction. Considerable effort has been directed toward single-video and multivideo super-resolution methods. We aim to obtain a higher registration accuracy that maximally employs video frames to reconstruct the current frame, particularly for moving or deforming objects. To this end, we provide a content-based registration algorithm based on a propagation matching algorithm and the Lucas–Kanade method. The super-resolution step is implemented using robust iterative minimization. We compare our algorithm to others and demonstrate that our algorithm achieves high registration accuracy and more effectively reconstructs fast moving and significantly deforming objects.
At present, in the field of image processing, the main algorithm to restore the blurred image is the blind deconvolution. But most of the blind deconvolution methods have to iterate a large amount of times and the result is also unsatisfactory. In this paper, a new blind deconvolution algorithm is proposed, which, consisting of two steps, is based on simultaneous estimating the specimen function and the parameters of the point-spread function (PSF). Firstly, it uses the expectation maximization algorithm (EM) to iterate the specimen function; secondly it uses the conjugate gradient method to estimate the parameters of the PSF. The mathematical model ensures that all the constraints of the PSF are satisfied, and the maximum-likelihood approach ensures that the specimen is nonnegative. In this paper, the general Gauss function is used to be as the PSF. In the experiment, it can successfully restore both the two-dimensional and three-dimensional images within limited times of iteration.
An algorithm for maximum-likelihood image restoration based on the expectation maximization (EM) algorithm is proposed in this paper. This estimation is based on a depth-variant imaging model in three-dimensional optical sectioning microscopy. As a result of the refractive index mismatch between the immersion medium and the mounting medium of the specimen, the imaging model in three-dimensional optical-sectioning microscopy incorporates spherical aberration that worsens with increasing depth under the coverslip and changes in the point spread function (PSF). Two-dimension images restoration and three-dimension serial images restoration are to be used to analyze the capability of the EM-ML algorithm, and the performance shows that the EM-ML algorithm can restore the blurred of image by the depth variant image model.
Proc. SPIE. 4922, Color Science and Imaging Technologies
KEYWORDS: Detection and tracking algorithms, Image segmentation, Video, Video surveillance, Surveillance, Signal processing, Information technology, Video processing, Optical character recognition, Fuzzy logic
A new method for the car plate extraction and recognition is presented in this paper. First, the segmentation technique, based on multiple threshold values determined by fuzzy entropy and the prior knowledge on car plate character, is used to extract the car plates, which has strong anti-noise capability, and is able to locate the car plates quickly in the varying backgrounds. Then, a recognition method combining the several recognizers is proposed to recognize the car plate characters. The recognition rates of over 97% under various illumination conditions in real applications shows that the proposed method is effective and reliable for car plate recognition.