Semantic segmentation is essential in medical image analysis. In this paper, we propose a lightweight vision transformer-based semantic segmentation method for medical images. The proposed method is composed of three parts: encoder, decoder, and segmentation head. The encoder is a light-weight ViT called MPiFormer developed based on iFormer and the split-and-merge strategy. A decoder is designed to fuse the multi-scale feature maps and a segmentation head is developed to predict the semantic segmentation map. Experiments on Barrett’s esophagus and Melanoma datasets demonstrate that the proposed method achieves an F1-score of 96.01% and an mIoU of 92.24% with limited annotations. Additionally, the split-and-merge mechanism reduces parameters and computation by 72.63% and 71.75%, respectively, while maintaining stable performance.

In this paper, we use Deep Convolutional Generative Adversarial Networks (DCGANs) method to generate more images with multiple views to increase our dataset diversity. We use 3D-model different views for training DCGAN to make interpolation between the leftest and rightest random vectors, which means it can generate leftest to rightest images. After producing many of multi-view images, we combine with CNN based modules called co-attention map generator to look for common features of the same class but in different views clothing. By applying the learned generator to all images, the corresponding co-attention maps are obtained. we can fluently apply the proposed method can function well for multi-view objects on different types of clothing classes.

In this paper, a vision-based video synchronization method was developed based on feature point matching. The proposed method contains several parts: pre-processing, key-frame matching, and frame synchronization. Since videos may have different frame rates and frame sizes, the pre-processing procedure is used to normalize the frame rate and frame size of videos. After key frame extraction, matching similar key frames from difference videos is conducted based on feature point matching and RANSAC. The frame index differences among videos can be obtained by analyzing these matched key frames and then synchronizing videos can be achieved. To evaluate the performance of the proposed method, some videos are captured for testing. Experimental results show that key frames can be selected effectively. Frame matching can be well achieved based on feature point matching and RANSAC. The higher score of subjective testing demonstrates that the proposed system can synchronize videos well.

In this paper, a helicobacter pylori classification method based on deep neural network (Inception v3) was proposed. The purpose of the proposed model is to provide physicians with reference to the diagnosis of Helicobacter pylori infection for increasing the diagnostic efficiency. Data augmentation and transfer learning are exploited for model construction to generate a classification system with high prediction accuracy. To evaluate the performance of the proposed method, many endoscope images are collected for testing. Experimental results show that the proposed method can well determine whether the input image contains Helicobacter pylori or not.

In this paper, we propose a new depth image-based rending scheme for 3DTV applications, where a background sprite model is utilized for dis-occlusion/hole filling purpose. Dissimilar to traditional spatial (e.g., interpolation or inpainting) and temporal methods, our algorithm is capable of recovering the holes with true background information by incrementally integrating the spatial and temporal information of the video in a unified background sprite model. The technique of background sprite model construction in this paper is featured of resolving camera motions existing in most of the consumer videos, accurate registration of multiframe information, and efficient memory use and computation in realistic 3DTV application. To register/stitch each input frame to the background sprite model accurately, foreground removal considering color and depth information, and an adaptive key-frame-based scheme are developed for transform computation. Experimental results show that our proposed scheme has a large temporal reference distance and can retrieve true background information accurately, thus leading to better quality after novel view synthesis compared to existing spatial or spatio-temporal algorithms, especially for videos with significant camera motions or complex backgrounds.

In this paper, we propose a new depth image-based rending (DIBR) scheme for 3DTV applications, which is based on temporal hole filling with sprite generation. Dissimilar to traditional methods (e.g., spatial interpolation and inpainting), temporal information is utilized to recover the holes with true background information. The proposed scheme is composed of two parts: sprite generation and virtual view synthesis. To collect sufficiently temporal information, a key-frame-based sprite generation algorithm was developed to incrementally and accurately fuse the background color and depth information from successive frames. Then the holes in the synthesized virtual view can be recovered by referring to the constructed sprite models. Experiments demonstrate that our proposed scheme is capable of achieving good results and outperforms some existing methods subjectively and objectively.

Breast cancer is one of the leading causes of death from cancer in Taiwan. In this paper, we propose a feature-based scheme composed of preprocessing, feature extraction and a fuzzy classifier for suspicious region detection and identification. In the preprocessing stage, we first extract regions of interest and then coarsely determine suspicious regions via candidate screening. Some features are extracted based on intra-slice, texture and inter-slice analysis techniques for suspicious region identification. Intra-slice analysis evaluates the intensity and size of suspicious regions. To find a precise region, we propose a region growing algorithm based on ellipse-based approximation. In texture analysis, some texture cues are extracted from spatial and wavelet domains and integrated as a combined texture feature by using a neural network. Inter-slice analysis is based on the continuity characteristic and consistency of a suspicious region's size; the objective is to verify the static behavior of suspicious regions. Several magnetic resonance imaging (MRI) cases are utilized to evaluate the performance of the proposed scheme. Experimental results demonstrate that our scheme can not only extract regions of interest but also identify tumors well from magnetic resonance images.

In this paper, we propose a new two-pass bi-directional keyframe depth propagation algorithm for semi-automatic 2D-to-3D video conversion. First, key-frames are selected from each video shot based on color compensation errors. Depths for key-frames are then manually assigned or rendered by using popular computer tools, which are then propagated to non-key-frames bounded by a pair of front and rear key-frames. Our two-pass bidirectional procedure is advantageous in solving the background occlusion/dis-occlusion problems that degrade traditional depth propagation Experiments show that our algorithm is capable of achieving better results in both subjective and objective evaluations.

In this paper, we present a quality enhancement scheme for endoscopic images. Traditional algorithms might be able to enhance the image contrast, but possible over-enhancement also lead to bad overall visual quality which prevents surgeons from accurate examination or operations of instruments in Minimal Invasive Surgery (MIS). Our proposed scheme integrates the well-known retinex algorithm with a pseudo-HDR (High Dynamic Range) synthesis process, designed to compose of three parts: multiscale retinex with gamma correction (MSR-G), local brightness range expansion (brightness diversity), and bilateral-filter-based HDR image fusion. Experiment results demonstrate that the proposed scheme is able to enhance image details and keep the overall visual quality good as well, with respect to other existing methods.

In this paper, we present a passive-blind scheme for detection of frame duplication forgery in videos. The scheme is a coarse-to-fine approach that is implemented in four stages: candidate segment selection, spatial similarity measurement, frame duplication classification, and post-processing. To screen and select duplicated candidates in the temporal domain, the histogram difference of two adjacent frames in the RGB color space is adopted as a feature. Then, to evaluate the similarity of two images, we use a block-based algorithm to measure the spatial correlation between the candidate segment and the corresponding frame in the query template. Based on the results of spatial and temporal analysis, we construct a classifier to detect duplicated clips. In addition, to deal with the partial detection problem, we develop a post-processing technique that examines and merges two adjacent detected candidates into a complete duplicated video clip. Our experiment results demonstrate that the proposed scheme can not only achieve detection of frame duplication forgery but also accurately detect and localize duplicated clips in different kinds of videos. The results also show that the scheme outperforms an existing method in terms of precision, recall, accuracy, and computation time.

In this paper, we propose an adaptive interpolation scheme based on iterative back-projection and human visual system based quality metric for image sequences. Initial estimates of each up-sampled image can be generated individually by using subpixel interpolation and subpixel motion estimation in the spatial and temporal domains respectively. Then, based on the initial estimates and edge information, up-sampled images are derived by using an iterative back-projection technique and a quality metric based on the human visual system. After fusing the up-sampled images into a final version, a low-pass filter is applied as a post-processing step to reduce the effect of blocking artifacts in each reconstructed up-sampled image. Our experimental results demonstrate that, in terms of PSNR (Peak Signal-to-Noise Ratio) and NQM (Noise Quality Metric), the proposed scheme outperforms four existing methods.

In this paper, we propose a passive-blind scheme for detecting forged images. The scheme leverages quantization table estimation to measure the inconsistency among images. To improve the accuracy of the estimation process, each AC DCT coefficient is first classified into a specific type; then the corresponding quantization step size is measured adaptively from its energy density spectrum (EDS) and the EDS's Fourier transform. The proposed content-adaptive quantization table estimation scheme is comprised of three phases: pre-screening, candidate region selection, and tampered region identification. In the pre-screening phase, we determine whether an input image has been JPEG compressed, and count the number of quantization steps whose size is equal to one. To select candidate regions for estimating the quantization table, we devise a candidate region selection algorithm based on seed region generation and region growing. First, the seed region generation operation finds a suitable region by removing suspect regions, after which the selected seed region is merged with other suitable regions to form a candidate region. To avoid merging suspect regions, a candidate region refinement operation is performed in the region growing step. After estimating the quantization table from the candidate region, an maximum-likelihood-ratio classifier exploits the inconsistency of the quantization table to identify tampered regions block by block. To evaluate the scheme's performance in terms of tampering detection, three common forgery techniques, copy-paste tampering, inpainting, and composite tampering, are used. Experiment results demonstrate that the proposed scheme can estimate quantization tables and identify tampered regions effectively.

In this article, we propose a 2D to 3D video conversion scheme for MPEG videos. The difficulty for 2D/3D conversion problem lies on depth estimation/ assignment with insufficient information. Our depth assignment is based on the analyses of multiple cues, e.g., motion parallax, atmospheric perspective, texture gradient, linear perspective, and relative height, separately for the foreground objects and the background area. To fit more kinds of videos, the proposed depth assignment scheme is content-adaptive by segmenting a video into shots and classifying each of them into three categories for different conversion schemes. Subjective experiments show that the 3D stereo video generated by using our depth assignment scheme and the Depth Image Based Rendering (DIBR) technique presents little difference to that created based on the depth ground truths.

In this paper, we proposed a passive scheme to achieve image tampering detection. To correctly estimate the quantization table, each AC DCT coefficient is classified into different types and then the corresponding quantization stepsize is adaptively measured from its power spectrum density (PSD) and PSD's Fourier transform. Based on the quantization table estimation, the proposed scheme is composed of pre-screening, candidate region selection, and tampering region identification. The pre-screening is developed to decide whether an image had been JPEG compressed. To select candidate regions for estimating quantization table, a seed region is first selected by finding a suitable region by removing suspicious regions. Based on the seed region, the candidate region can be obtained by suitably merging other regions into the seed. To avoid merging the suspect regions, a candidate region refinement is performed in the region growing. After estimating the quantization table from the candidate region, a MLR classifier based on the inconsistency of quantization table is exploited to identify tampered regions block by block. The experimental results demonstrate that our proposed scheme can not only estimate the quantization table but also identify tampered regions well.

Picture quality and statistical undetectability are two key issues related to steganography techniques. In this paper, we propose a closed-loop computing framework that iteratively searches proper modifications of pixels/coefficients to enhance a base steganographic scheme with optimized picture quality and higher anti-steganalysis capability. To achieve this goal, an anti-steganalysis tester and an embedding controller-based on the simulated annealing (SA) algorithm with a proper cost function-are incorporated into the processing loop to conduct the convergence of searches. The cost function integrates several performance indices, namely, the mean square error, the human visual system (HVS) deviation, and the differences in statistical features, and guides a proper direction of searches during SA optimization. Our proposed framework is suitable for the kind of steganographic schemes that spreads each message information into multiple pixels/coefficients. We have selected two base steganographic schemes for implementation to show the applicability of the proposed framework. Experiment results show that the base schemes can be enhanced with better performances in image PSNR (by more than 5.0 dB), file-size variation, and anti-steganalysis pass-rate (by about 10% ~ 86%, at middle to high embedding capacities).

In this paper, an optical cryptosystem with adaptive steganography is proposed for color image encryption and decryption. The optical cryptosystem employs a double random phase encoding algorithm to encrypt and decrypt color images. The color image is first separated into three channels: red, green, and blue. Each channel is encrypted by two random phase masks generated from session keys. For higher security, an asymmetric method is applied to cipher these session keys. The ciphered data produced by the asymmetric method is then embedded into the encrypted color image by a content-dependent and low distortion data embedding technique. The key delivery is accomplished by hiding ciphered data into the encrypted color image with a specific hiding sequence generated by the zero-LSB sorting technique. Experimental results show that the proposed adaptive steganography-based cryptosystem has a good performance when it is applied to color images.

In this paper, we propose a feature-based scheme for detecting different genres of video shot transitions based on spatio-temporal analysis and model parameter estimation. We use the histogram difference and a modified version of it to detect cuts and reduce the impact of fleeting lights. To deal with dissolves, a hybrid algorithm composed of adaptive thresholding, parameter calculation, and transition duration refinement is proposed. In addition, to find a precise duration, we develop a transition duration refinement method to examine each dissolve candidate. The associated model parameters of the dissolve are also estimated as features. After feature extraction, a fuzzy classifier integrates the features to distinguish non-transitions, cuts, and dissolves from one to another. The results of experiments demonstrate that the proposed scheme not only detects cuts and dissolves well, but also accurately locates the duration of each dissolve transition. In addition, the proposed scheme outperforms existing methods in terms of cut and dissolve detection.

In this paper, we propose a feature-based scheme for detecting different genres of video shot transitions based on spatio-temporal analysis and model parameter estimation. In feature extraction, the histogram difference and its modified versions are calculated from the effectiveness of detecting cuts and reducing the impact of fleeting lights. We propose a hybrid algorithm composed of adaptive thresholding, parameter calculation, and transition duration refinement to measure model parameters. Some properties of the associated model parameters of each transition are computed as features. A feature measuring the time gap between two consecutive shots is also adopted. After feature extraction, a fuzzy classifier integrates these features to distinguish nontransitions, cuts, and dissolve-type features from one to another. Many test videos having different types of shots are used for performance evaluation. The experimental results demonstrate that the proposed scheme not only detects cuts, dissolves, and fades well, but also accurately locates the duration of each dissolve-type transition. In addition, the proposed scheme outperforms some existing methods in terms of cut and dissolve detection.

In this paper, we propose an architecture that combines a quantization-based steganogrphic scheme with a steganalysis system, operated in a closed-loop manner with a cost function for minimization, to enhance the anti-steganalysis capability and image quality after data embedding. In this architecture, a controller based on the SA (simulated annealing) technique is adopted to guarantee fast and accurate convergence. Our proposed system is applied to the data embedding of JPEG-compressed images. Compared with the original embedding algorithm in (Seki, Y, et al., 2005), a better image quality (by an average improvement of 6.48 dB) can be achieved and simultaneously the anti-steganalysis capability is enhanced significantly.

In this paper, the authors propose a watermarking scheme that embeds both image-dependent and fixed-part marks for dual protection (content authentication and copyright claim) of JPEG images. To achieve the goals of efficiency, imperceptibility, and robustness, a compressed-domain informed embedding algorithm, which incorporates the Lagrangian multiplier optimization approach and an adjustment procedure, is developed. A two-stage watermark extraction procedure is devised to achieve the functionality of dual protection. In the first stage, the semifragile watermark in each local channel is extracted for content authentication. Then, in the second stage, a weighted soft-decision decoder, which weights the signal detected in each channel according to the estimated channel condition, is used to improve the recovery rate of the fixed-part watermark for copyright protection. The experiment results manifest that the proposed scheme not only achieve dual protection of the image content, but also maintain higher visual quality (an average of 6.69 dB better than a comparable approach) for a specified level of watermark robustness. In addition, the overall computing load is low enough to be practical in real-time applications.

In this paper, a prescription-based error concealment (PEC) method is proposed. PEC relies on pre-analyses of the concealment error image (CEI) for I-frames and the optimal error concealment scheme for P-frames at encoder side. CEI is used to enhance the image quality at decoder side after error concealment (by spatial interpolation or zero motion) of the corrupted intra-coded MBs. A set of pre-selected error concealment methods is evaluated for each corrupted inter-coded MB to determine the optimal one for the decoder. Both the CEI and the scheme indices are considered as the prescriptions for decoder and transmitted along with the video bit stream based on a reversible data embedding technique. Experiments show that the proposed method is capable of achieving PSNR improvement of up to 1.48 dB, at a considerable bit-rate, when the packet loss rate is 20%

In contrast to steganography, steganalysis is focused on detecting (the main goal of this research), tracking, extracting, and modifying secret messages transmitted through a covert channel. In this paper, a feature classification technique, based on the analysis of two statistical properties in the spatial and DCT domains, is proposed to blindly (i.e., without knowledge of the steganographic schemes) to determine the existence of hidden messages in an image. To be effective in class separation, the nonlinear neural classifier was adopted. For evaluation, a database composed of 2088 plain and stego images (generated by using six different embedding schemes) was established. Based on this database, extensive experiments were conducted to prove the feasibility and diversity of our proposed system. It was found that the proposed system consists of: 1) a 90%/sup +/ positive-detection rate; 2) not limited to the detection of a particular steganographic scheme; 3) capable of detecting stego images with an embedding rate as low as 0.01 bpp; and 4) considering the test of plain images incurred low-pass filtering, sharpening, and JPEG compression.

A statistical feature-based scheme is proposed to identify the data hiding domain of an embedded signal in this research. Two phenomena are observed for images before and after data hiding. First, the gradient energy increases as the continuity of gray levels between adjacent pixels is disturbed by the embedded signal. Second, the statistical variance of the coefficient distribution in macro-blocks tends to decrease after data hiding. These phenomena are analyzed mathematically. Then, statistical features in the pixel, DCT, and DWT domains are extracted and a maximum likelihood ratio test is adopted to solve the hiding domain classification problem. The proposed scheme has demonstrated good classification results.

Fragile watermarking is designed to protect content integrity of digital data, meaning to detect any change in the content as well as localize areas that have been changed. In this paper, we propose a fragile watermarking scheme, which embeds dual watermarks (one is naturally fragile and the other is actually robust) in JPEG-2000 compressed images with ROI (region of interest) selection. In this scheme, the fragile watermark is embedded in ROI coefficients at the first decomposition layer, while the robust watermark embedded in ROB (region of backgrounds) coefficients at the third decomposition layer. The combinational working of them is not only capable of detecting malicious changes precisely and flexibly but also estimating the degree of alteration to discriminate intended attacks from unintended ones (e.g., such normal image processing as compression, low-pass filtering, sharpening, etc). Experiments show that our functionally fragile watermarking scheme is scalable in view of detecting attack's intension as well as accuracy in localizing the tampered areas.

A new public key crypto-system for transmitting secured MPEG videos is proposed in this paper. Our proposed system retains the MPEG bit stream structure, but with the DC differential values and some AC coefficients of intra-coded macroblocks being encrypted, hence resulting in a content invisibility for all I-, P-, and B-frames. In this way, our algorithm is efficient in terms of processing speed and data conversion overhead. Parameters that control the generation of several pseudo-random binary patterns for encryption can be dynamically updated, encrypted by a public-key algorithm, and delivered to the receiver through the encrypted MPEG video by data embedding techniques. Since these parameters can be recovered for legal users by using a private key, transmission security can be thus achieved. Besides, our crypto-system is capable of changing these parameters dynamically and making them readable at the receiver side, it is difficult for illegal users to deduce the decryption parameters by some statistical methods.

This paper proposes two simple DCT-domain-based schemes to embed single or multiple watermarks into an image for copyright protection and data monitoring and tracking. The watermark data are essentially embedded in the middle band of the DCT domain to make a tradeoff between visual degradation and robustness. The proposed schemes are simple and no original host image is required for watermark extraction. The algorithm also features the capability of embedding multiple orthogonal watermarks into an image simultaneously. A set of systematic experiments, including Gaussian smoothing, JPEG compression, and image cropping are performed to prove the robustness of our algorithms.