基于人眼视觉特性的自适应支持权重立体匹配算法

Stereo matching is one of the research directions of binocular stereo vision system. In this paper, a novel adaptive local stereo matching algorithm is proposed. Firstly, the cost function calculation is redefined according to the gray similarity and the spatial similarity. Then, the data smoothing term is introduced and the penalty coefficient is determined by the local entropy of the image. Finally, the initial disparity map of the cost aggregation is analyzed by using the left-right consistency check and weight median filter to perform disparity refinement. The algorithm uses four international standard stereo image pairs provided by Middlebury website. The experiment shows that the proposed algorithm can get more accurate disparity map.

Medical image enhancement is to process the image so that the result is more suitable than the original image for analysis and diagnosis. The information of image is acquired by the observer's eyes, so in the image processing, the human visual characteristics (HVC) as an important factor should be well thought of. In the paper, we study HVC from testing the distinguishability of gray scale and the sensitivity to the structure of image, analyze the physiological character and psychological characters of human vision, and summarize the laws of HVC. A new medical image enhancement method is presented based on HVC. It divides the image into the smooth areas and detail areas, and uses different processing methods for the areas according to HVC. Experiment results demonstrate that the method enhances the edge detail information, and improves the image quality, thereby increases the veracity of diagnoses, makes the treatment approach more exact.

As to the problem of great complexity of traditional adaptive window matching algorithm, a simple adaptive window matching algorithm is proposed in this paper, which use various square or rectangular windows of different size based on the different gradient of points to be matched. Moreover, for better matching effect, a regional seed propagation method is proposed based on the disparity smoothness constraints and the assumption that adjacent pixels of similar color should have the same disparity. Experiments show that compared with traditional adaptive window matching algorithms, the computation complexity has been decreased significantly and the matching accuracy rate has been improved effectively.

In this paper, we propose an adaptive stereo matching algorithm to encompass stereo matching problems in projective distortion region. Since the projective distortion region can not be estimated in terms of fixed-size block matching algorithm, we tried to use adaptive window warping method in hierarchical matching process to compensate the perspective distortions. In addition, probability theory was adopted to encompass uncertainty of disparity of points over the window in this study. The proposed stereo matching algorithm has been tested on both disparity map and 3D model view. The experimental results show that remarkable improvement is obtained in the projective distortion region.

Dense feature extraction is becoming increasingly popular in face recognition tasks. Systems based on this approach have demonstrated impressive performance in a range of challenging scenarios. However, improvements in discriminative power come at a computational cost and with a risk of over-fitting. In this paper, we propose a new approach to dense feature extraction for face recognition, which consists of two steps. First, an encoding scheme is devised that compresses high-dimensional dense features into a compact representation by maximizing the intrauser correlation. Second, we develop an adaptive feature matching algorithm for effective classification. This matching method, in contrast to the previous methods, constructs and chooses a small subset of training samples for adaptive matching, resulting in further performance gains. Experiments using several challenging face databases, including labeled Faces in the Wild data set, Morph Album 2, CUHK optical-infrared, and FERET, demonstrate that the proposed approach consistently outperforms the current state of the art.

HVS theory plays important role in the application of digital image watermarking technique. When inserting watermarking, the visual masking feature of HVS could be fully used to design digital watermarking algorithm with good perceived performance. When extracting watermarking from the damaged image, human's visual feature could be combined to recover the damaged image so as to obtain better effect. This article combines with human visual feature to study digital watermarking technique, utilizes watermarking redundant technique to insert watermarking, extracts watermarking in line with the damage situation of watermarking, and combines with visual redundancy feature to achieve a image scrambling algorithm that is easy to recover and a recovery scheme for damaged scrambling image.

The online delivery of physical education poses challenges due to its distinctive attributes, primarily stemming from the difficulty in ensuring comprehensive oversight of instructional content. In response, we advocate for the implementation of a multi-faceted computational system within a big data framework tailored explicitly for online physical education. Initially, we engineer a parallel, multi-objective real-time instructional supervision mechanism employing sensor technology and long-range transmission capabilities to cultivate an environment conducive to physical education while efficiently gathering pertinent data. Subsequently, we introduce a multi-functional robot-assisted methodology for supervising physical education intelligently, proficiently overseeing students’ activities and sports engagements, thereby facilitating comprehensive analysis of instructional content. Concluding this framework, we present an adaptive matching algorithm tailored for physical education instruction, utilizing the outcomes of content analysis and sports item recognition to recommend subsequent instructional segments. Our demonstration validates the precision of our teaching assistant methodology in accurately identifying sports items, achieving an impressive accuracy rate of 90.72%. Furthermore, our adaptive matching algorithm garners an 80% approval rating in subjective evaluations, establishing its utility as a supplemental resource for online physical education.

We have developed several three-dimensional display systems that are matched to the human visual field characteristics. In this article, we describe our developed display systems, which are matched for the human communication in the closerange, medium-range, and distant rage categories.

In this paper, we propose a stereo matching method based on adaptive plane-fitting. In the proposed method, we use the disparity map adjustment with adaptive weight filter to preserve the boundary region. To estimate the initial disparity map, we use two cost aggregation functions, and in addition to that, we combined color and gradient for cost computation. In addition, the error value of the initial disparity map is calibrated by applying plane-fitting using mean-shift segmentation. The disparity map is adjusted by using adaptive weights to improve the accuracy of the disparity map. Experimental results show that the proposed method generates a high-quality disparity map which has the distinct boundary between objects compared to other methods.

Surveillance video cameras capture large amounts of continuous video streams every day. To analyze or investigate any significant events, it is a laborious and boring job to identify these events from the huge video data if it is done manually. Existing approaches sometimes neglect key frames with significant visual contents and/or select some unimportant frames with low/no activity. To solve this problem, in this paper, a video summarization technique is proposed by combining three multimodal human visual sensitive features, such as foreground objects, motion information, and visual saliency. In a video stream, foreground objects are one of the most important pieces of a video as they contain more detailed information and play a major role in important events. Moreover, motion is another stimulus of a video that significantly attracts human visual attention. To obtain this, motion information is calculated in the spatial domain as well as the frequency domain. Spatial motion information can select object motion accurately; however, it is sensitive to illumination changes. On the other hand, frequency motion information is robust to illumination change, although it is easily affected by noise. Therefore, motion information in both the spatial and the frequency domains is employed. Furthermore, the visual attention cue is a sensitive feature to measure the indication of a user's attraction label for determining key frames. As these features individually cannot perform very well, they are combined to obtain better results. For this purpose, an adaptive linear weighted fusion scheme is proposed to combine the features to rank video frames for summarization. Experimental results reveal that the proposed method outperforms the state-of-the-art methods.

ADS40 is the first commercial airborne pushbroom sensor system produced by LH Systems, which offers on a single camera system the possibility to acquire both 3-line panchromatic and multispectral images in up to 10 channels. The geometric configuration and characteristics of the sensor enable the development of a matching algorithm, based on a multi-image philosophy, incorporating novel strategies, primarily aiming at producing reliable DSMs/DTMs. Existing algorithms are geared towards frame imagery and do not make explicit use of the special characteristics of the sensor; thus, most of them show a low success rate, requiring additional manual editing. Multi-image constraints make use of quasi-epipolar lines, since each line has its own position and attitude, to restrict search space. A multi-patch approach, utilized in the algorithm, strengthens the identification of mismatches and can be justified since larger patches aim at reliability of a coarse solution and the smaller ones at accuracy. Additional quality measures, computed at run time, are combined in the stage of the quality control in order to assign each matched point to a possible pre-defined error class (e.g. occlusion, multiple solutions). The evaluation of the matching performance is done through a qualitative and quantitative analysis using manually measured mass points from a block of frame aerial images with the same ground sampling distance (0.22 m) as ADS40 data.

Product packaging design is a work that integrates graphic design and three-dimensional design. 3D visualization technology and digital virtual interaction technology in virtual reality are developing rapidly in packaging design applications. On the one hand, this study discusses the application of digital technology in the three-dimensional design of packaging structure, including packaging carton structure design, packaging container modeling design, and packaging two-dimensional visual design. On the other hand, this study studies the color correction algorithm of packaging design and proposes a color correction based on the visual characteristics of human eyes. The pattern search algorithm finds a local optimum, and the initial value affects the correction result. Therefore, inspired by the color perception properties of the human visual system, the image is corrected by combining visual multichannel and visual nonlinearity. The experimental results of the algorithm are more consistent with human visual perception.

An adaptive Fuzzy Inference Perceptual model has been proposed for watermarking of video streams. The model is designed to be adaptive to the human visual characteristics to attest the owner identification and discourage the unauthorized copying. The proposed model integrates the human visual characteristics of video motion sub-regions, in the frequency multi-resolution wavelet domain, with multi-dimensional fuzzy inference perceptual model. The designed fuzzy multi-variable input system inherits luminance, texture and gradient to generate an adaptive watermark embedding strength factor. Interacting motion estimation is applied for the selection of the candidate frames and sub-blocks. A fuzzy-based scheme has been proposed to generate a perceptual membership degree for strength watermark embedding factor for the selected candidate motion blocks. Experiments have been carried out with different benchmark test videos of different sizes, visual characteristics and sampling rates. Various sizes of watermark signatures have been applied on the model. Several experimental attacks have been applied such as frame dropping, frame averaging, JPEG compression, Gaussian noises. It manifests considerable robustness against various geometric and signal processing attacks. Several attacks have been applied to the proposed scheme and the experiments revealed promising results in terms of visual quality and extracted watermark distortion. In addition, the model has been compared with different other watermarking schemes in literature. The proposed model showed superior performance in terms of fidelity, robustness to attacks and high level of imperceptibility.

The optical fiber distributed strain sensor based on the optical frequency domain reflectometer (OFDR) preserves its dominant position in short-distance measurement fields with high spatial resolution, such as biomedical treatment, soft robot, etc. However, owing to the weak intensity of the Rayleigh backscattered signal (RBS) in the single-mode fiber (SMF) and complex computation, the large strain changes cannot be precisely and rapidly demodulated by the traditional cross-correlation method. In this work, the OFDR with backscattering enhanced optical fiber (BEOF) is proposed and demonstrated for fast and large strain measurement. By enhancing the RBS amplitude, the signal-to-noise ratio (SNR) is improved, resulting in a higher similarity between the reference signal and test signal, which is beneficial for the expansion of the strain measurement range. Moreover, the adaptive local feature extraction and matching (ALFEM) algorithm is presented and demonstrated, which replaces the traditional cross-correlation method for strain demodulation and fast measurement. On account of the enhancement ratio of BEOF, the dominant characteristic data segment can be extracted from whole wavelength data. In the experiments, the enhancing ratio of BEOF is designed as 10, resulting in the spatial resolution reaches 400µm and the strain measurement range is greatly increased to 4800µɛ. Further, the effectiveness of the ALFEM algorithm has been verified, in which the strain demodulation time is approximately 25% of that of the traditional method. This scheme fully exploits the enhancement characteristic of the BEOF and is also applicable to the systems based on other types of BEOF, different strain changes and sensing distances.

In video communication, the quality of video is mainly determined by bitrate in general. Moreover, the effect of video contents and their visual perception on video quality assessment (VQA) is often overlooked. However, in fact, for different videos, although the bitrates are the same, their VQA scores are still significantly different. Hence, it is assumed that the bitrate, video contents, and human visual characteristics mainly affect the VQA. Based on the above three aspects, in this paper, we designed a bitrate-based no-reference (NR) VQA metric combining the visual perception of video contents, namely, BRVPVC. In this metric, first an initial VQA model was proposed by only considering the bitrate alone. Then, the visual perception model for video contents was designed based on the texture complexity and local contrast of image, temporal information of video, and their visual perception features. Finally, two models were synthesized by adding certain weight coefficients into an overall VQA metric, namely, BRVPVC. Furthermore, ten reference videos and 150 distorted videos in the LIVE video database were used to test the metric. Moreover, based on the results of evaluating the videos in LIVE, VQEG, IRCCyN, EPFL-PoliMI, IVP, CSIQ, and Lisbon databases, the performance of BRVPVC is respectively compared with that of six full-reference (FR) metrics and ten NR VQA metrics. The results show that our VQA metric has a higher accuracy than six common FR VQA metrics and eight NR VQA metrics, and it is close to other two NR VQA metrics in accuracy. The corresponding values of Pearson linear correlation coefficient and Spearman rank order correlation coefficient reached 0.8547 and 0.8260, respectively. In addition, the computational complexity of proposed VQA metric is lower than video signal-to-noise ratio, video quality model, motion-based video integrity evaluation, spatiotemporal most apparent distortion, V-BLINDS, and V-CORNIA metrics. Moreover, the proposed metric has a better generalization property than these metrics.