Search

An important goal of color image gray-scale is to keep the edge details of the original color image as much as possible. In many cases, the degree of feature discrimination is maintained, but in some cases, edge details are still lost or blurred. Therefore, this paper first uses an improved non-linear global mapping grayscale method to grayscale the color image, and then proposes a grayscale image detail enhancement algorithm based on rolling guided filtering. The method in this paper is to enhance the edge details of the grayscale image by rolling guided filter processing on the basis of the grayscale image. In addition, the rolling-guided filter is a local linear model with better edge retention characteristics, which can overcome the defect that other filters are prone to gradient flips on the edges where the gray level of the image changes sharply, causing the image to appear “false edges”. The experimental results show that when the traditional method loses or blurs the detailed features, the method in this paper can maintain better detailed features.

Contrast enhancement is an important pre-processing task in any Image Analysis (IA) system. In this paper, we formulate the image contrast enhancement problem as an optimization problem where the goal is to optimize the pixel intensity values of an input image to obtain a contrast enhanced version of the same. This optimization task is executed by suitably customizing a nature-inspired optimization algorithm called Selfish Herd Optimizer (SHO). The optimization problem is solved using two different solution representations: pixel wise optimization (SHO(direct)) and transformation function based optimization (SHO(transformation)). Moreover, an ablation study is performed to select the most appropriate parameters which can be used in fitness measure for this optimization problem. On experimenting over the popular Kodak image dataset, it has been observed that the proposed methods outperform many existing methods published recently. Further comparisons indicate that the direct approach performs better than its transformation counterpart. This paper further investigates the robustness of SHO(direct) approach by applying it to enhance the degraded document images of H-DIBCO 2018.

A novel color image encryption algorithm based on the fractional order laser chaotic system and DNA mutation principle is proposed in this paper. Using phase diagram, Lyapunov exponential spectrum, bifurcation diagram and C0 complexity, the dynamic characteristics of the fractional-order laser chaotic system are analyzed, and the hardware circuit simulation of the chaotic system is realized on the DSP platform. Based on the comprehensive analysis of the system, we have designed an image encryption algorithm. Firstly, the chaotic sequence and Arnold matrix are used to scramble the value of the image. Then, in the diffusion link, DNA diffusion algorithm and DNA mutation theory were used for the first time, which increased the randomness of the matrix. The performances of the designed encryption scheme are analyzed by key space, correlation coefficients, information entropy, histogram, differential attacks and robustness analysis., the experimental results prove that the algorithm has strong encryption capabilities and can withstand multiple decryption methods. The new color image encryption scheme proposed in this paper can realize the secure communication of digital images.

The usage of productive baseband receivers are portrayed by reasonable computational load is an urgent point in the advancement of transmission frameworks misusing assorted qualities in various spaces.Several research works have been published in the literature ofspace-time block codes multi-carrier code-division multiple-access (STBC MC-CDMA)Systems based on minimum conditional bit error rate Criterion. As like, in our previous paper, an efficient STBC communication system was performed using oppositional krill herd algorithm (OKHA). To overcome the drawbacks in the previous work, in this work we proposed a hybrid Particle swarm optimization (PSO) and group search optimization (GSO), called PSGSO based Multi-User Detection in STBC MC-CDMA Systems. In this paper, we use the Alamouti’s STBC encoder for transmitting sequence and the channel estimation matrix is optimally evaluated using PSGSO algorithm. The simulation results show that the proposed algorithm is properly reducing the bit error rate (BER), symbol error rate (SER) and properly increase the spectral efficiency value compare to other techniques.

Authenticity in multimedia information retrieval system is one such parameter which is of utmost in demand. One of such parameters is the biometric based authentication system, as seen in case of human the system proves its importance in almost each and every field. But in case of animals it becomes quite complex to recognize them using unique biometric parameter. This paper presents a modified 10 layered DenseNet deep learning framework which is implemented for the identification of Humpback whale, in an online challenge conducted by the kaggle, using biometric parameter. The proposed dense model minimizes the traditional limitation of vanishing gradient problem and outperforms as compared with other methods exists in the literature. This paper also proposed an hand free flow algorithm for ROI segmentation. In addition, the proposed methodology also uses CLAHE as a preprocessing method and focused on the feature re-usability with an optimized number of parameters used for recognition. During the experimentation, validation of the proposed model is found satisfactory with an accuracy of 93.21%. The test set results are provided by the kaggle with an AUC-ROC result over public and private scoreboard with an accuracy of 94.47% and 92.54% respectively. The comparative result with other deep methodologies suggests that the highest accuracy gained is 95.55% by Se-ResNet50 by some other learderboard and proposed model is found at second place.

This paper proposes a new rigid registration algorithm based on the rational quadratic kernel to align point sets with outliers and noise. First of all, the multi-source point sets may contain a lot of outliers and noise and the traditional registration algorithm cannot handle the outliers and noise efficiently, this paper introduces the rational quadratic kernel to the rigid registration problem, which can resist outliers and suppress noise to improve the registration accuracy. Secondly, based on the new registration model, we present an iterative closest point (ICP) algorithm and use Lagrange multiplier and the singular value decomposition (SVD) to compute the rigid transformation. Moreover, the effect of the parameter is discussed detailly and a useful parameter control method is introduced to increase the accuracy and robustness of registration. A series of experiments on simulations and real data demonstrate that the proposed algorithm is more precise and robust than other algorithms.

Image segmentation is the prime factor to elicit the detailed investigation of an image. The desired information from an image can be easily obtained through the intuitive technique called thresholding. In this technique, detailed analysis of different classes of an image is realised through Multilevel Thresholding (MLT). Accurate and ideal threshold values achieved by non-parametric objective functions such as Tsallis and Renyi are briefed in this paper. The non-additive property of Tsallis and entropic threshold selection property of Renyi drive to search the global threshold value precisely. Higher the segmentation level more is the computational time for exploration of optimal threshold with Tsallis and Renyi. This challenge is countered by MLT based Tsallis and Renyi, aided with Exchange Market Algorithm (EMA). Several research on nature- inspired algorithms such as Bacterial Foraging (BF), Particle Swarm Optimisation (PSO) and Genetic Algorithm (GA) are carried out. For the first time, this paper proposes a powerful metaheuristic EMA technique for image segmentation, which implements the strategies of shareholders in stable and unstable mode to earn profit. The cognizance of the shareholders is extracted to attain the desired goal to reach the global threshold avoiding premature convergence. Empirical outcome of the results indicate that outstanding tuning search is achieved by EMA compared to extensive search techniques such as PSO, BF and GA. Exploration and exploitation assessment by metrics such as stability, computational efficiency, Peak Signal to Noise Ratio (PSNR), uniformity measure and Wilcoxon rank sum test affirm the Tsallis and Renyi based EMA surpass the existing techniques to analyse the real-world images.

The crowding in bus is an important factor affecting passenger satisfaction and bus dispatching level. However, how to use video images to detect crowding accurately is a difficult problem. In this paper, firstly, an image sample library is established based on the evaluation standard of crowding in bus, which contains 16346 sample images. Then, Local Binary Pattern (LBP) and Gray Level Co-occurrence Matrix (GLCM) are used to extract the texture features of the image in bus. Then, a rough classification method of crowding based on Support Vector Machine (SVM) is proposed. At the same time, in order to improve the accuracy of rough classification of crowding, the optimization effects of grid search algorithm, particle swarm optimization algorithm and genetic algorithm on SVM parameters are compared. The results show that the optimization effect of genetic algorithm is the best, and the accuracy rate is 93.20%. Finally, for the problem that the SVM method is not ideal in the fine classification of crowding, this paper proposes a new method based on ResNet. SGD, Adadelta and Adam are selected to optimize the parameters of ResNet model. The accuracy of the optimal Adam algorithm reaches 96.22%, which effectively solves the problem of the fine classification of crowding in bus.

The semantic segmentation process divides an image into its constituent objects and background by assigning a corresponding class label to each pixel in the image. Semantic segmentation is an important area in computer vision with wide practical applications. The contemporary semantic segmentation approaches are primarily based on two types of deep neural networks architectures i.e., symmetric and asymmetric networks. Both types of networks consist of several layers of neurons which are arranged in two sections called encoder and decoder. The encoder section receives the input image and the decoder section outputs the segmented image. However, both sections in symmetric networks have the same number of layers and the number of neurons in an encoder layer is the same as that of the corresponding layer in the decoder section but asymmetric networks do not strictly follow such one-one correspondence between encoder and decoder layers. At the moment, SegNet and ESNet are the two leading state-of-the-art symmetric encoder-decoder deep neural network architectures. However, both architectures require extensive training for good generalization and need several hundred epochs for convergence. This paper aims to improve the convergence and enhance network generalization by introducing two novelties into the network training process. The first novelty is a weight initialization method and the second contribution is an adaptive mechanism for dynamic layer learning rate adjustment in training loop. The proposed initialization technique uses transfer learning to initialize the encoder section of the network, but for initialization of decoder section, the weights of the encoder section layers are copied to the corresponding layers of the decoder section. The second contribution of the paper is an adaptive layer learning rate method, wherein the learning rates of the encoder layers are updated based on a metric representing the difference between the probability distributions of the input images and encoder weights. Likewise, the learning rates of the decoder layers are updated based on the difference between the probability distributions of the output labels and decoder weights. Intensive empirical validation of the proposed approach shows significant improvement in terms of faster convergence and generalization.

Bus passenger flow information is very important as a reference data for bus company line optimization, schedule scheduling basis, and passenger travel mode arrangement. With the development of image processing technology, it has become a current research trend to count passenger flow with the help of surveillance video of passengers getting on and off the bus. The specific research contents of this paper based on video image detection and statistics of passengers are as follows:(1) Collect head target image samples through a variety of ways, including 3960 positive head target samples and 4150 negative head target samples, which together constitute the head target feature database. (2) Established a head target detection model based on deep learning. First, the labeling of the head target training data set is completed. Then, after 15,000 iterations of model training, the YOLOv3 head target detection network model was obtained, with a recall rate of 92.12% and an accuracy rate of 89.71%. (3) A multi-target matching tracking algorithm based on the combination of Cam-shift and YOLOv3 is proposed. First, the Cam-shift algorithm is used to track the head target. Secondly, the head target tracking data and the YOLOv3 detection data are combined to solve the problem of drift during the tracking of the Cam-shift algorithm through the data association matching method based on the minimum distance, and then combined with the time constraint, a passenger location information judgment rule is proposed. Optimize the error and missed detection in the process of head target detection and tracking, and improve the reliability of passenger trajectory tracking. (4) A statistical algorithm for the detection of passengers getting on and off the bus is proposed. First, the trajectory of passengers in the bus boarding and disembarking area is analyzed, and a process for judging passengers’ boarding and boarding behavior is proposed. At the same time, a passenger position information judgment rule is proposed according to the different situations of whether there are new passengers or missing passengers, so as to optimize the problem of wrong detection and missing detection in the process of head target detection and tracking. (5) Finally, experiments are carried out in actual bus scenes and simulation scenes. The experiment proves that the statistical algorithm for the detection of passengers getting on and off the bus proposed in this paper has good detection, tracking and statistics effects in bus scenes and simulation scenes.

This paper presents a novel encryption scheme based on genetic operations and Mixed Linear-Nonlinear Coupled Logistic Map Lattice (MLNCML) spatiotemporal chaotic system. The initial values of the MLNCML system are related to the plain-image. Therefore, this scheme is sensitive to the plaintext and keys. For the simplicity and efficiency of the encryption process, expanded XOR (eXOR) operation and genetic operations are applied in the scheme. Spatiotemporal chaotic system can resist dynamical degradation and periodic phenomena, which offers good randomness for encryption. Moreover, the eXOR operation in this paper can effectively avoid the chosen-plaintext attack of traditional XOR operation. The DNA mutation and crossover operations are controlled by chaotic sequences, which overcome man-in-the-middle attack in the traditional DNA addition and subtraction operation. The experimental results indicate that the proposed scheme has good security to resist common attacks.

In this paper we introduce a cavity reconstructing algorithm for 3D surface scans (CRASS) developed for filling cavities in point clouds representing human body surfaces. The presented method uses Bezier patches to reconstruct missing data. The source of input data for the algorithm was an 8-directional structured light scanning system for the human body. Typical 3D scan representing human body consists of about 1 million points with average sampling density of 1 mm. The paper describes the complete scan processing pipeline: data pre-processing, boundary selection, cavity extraction and reconstruction, and a post-processing step to smooth and resample resulting geometry. The developed algorithm was tested on simulated and scanned 3D input data. Quality assessment was made based on simulated cavities, reconstructed using presented method and compared to original 3D geometry. Additionally, comparison to the state-of-the-art screened Poisson method is presented. Values’ ranges of parameters influencing result of described method were estimated for sample scans and comprehensively discussed. The results of the quantitative assessment of the reconstruction were lower than 0,5 of average sampling density.

In this paper, a new scrambling algorithm is proposed to reduce the correlation between the image pixels in three directions. In this paper, this algorithm is called ladder scrambling, and the image is scrambled multiple times in the form of diagonal lines. In terms of diffusion, DNA coding is used, and the secondary processing of the image is performed through DNA coding to change the pixel value of the image. The final encrypted image is affected by three conditions: the number of scrambling algorithms, the selection of DNA coding rules, and the key of the chaotic system. Theoretical analysis and experimental results show that the algorithm is safe and effective.

Based on the n-ary counting system, combined with the matrix semi-tensor product theory and Hilbert curve, a chaotic image encryption algorithm is designed. Different from the traditional encryption method, the algorithm proposed in this paper is an encryption algorithm with scrambling and diffusion at the same time. First, the pixel value is converted from decimal to n-ary. In the n-ary counting system, the plaintext image is randomly divided into some groups, and the Hilbert curve is used for scrambling to each group. The blocks are converted into scrambled images, so that the scrambling and diffusion can be carried out at the same time. Then, in order to improve the security of the algorithm, another round of diffusion is carried out based on matrix semi-tensor product mechanism. Chaotic sequence is generated by Chen system. This chaotic sequence performs matrix semi-tensor product operation with the first round of encrypted image, and generate second encrypted images. Finally, this encryption method is applied to color image encryption. Compared with some representative algorithms, the experimental results show that the algorithm proposed in this paper is secure and it can resist common attacks.

In this paper, we propose a fast Frame Rate Up-Conversion (FRUC) method based on Quaternion Wavelet Transform (QWT) motion estimation to improve the motion estimation accuracy and reduce the computational complexity. The proposed method contains three key elements: motion estimation, motion post-processing, and motion compensated frame interpolation. This paper considers the QWT motion estimation using the phase of image sequences, which leads to more accurate motion estimation, less post-processing procedures, and low complexity as well. Moreover, we proposed a modified fuzzy vector mean filtering to cope with motion outliers and made use of bilateral motion vector modification in the post-processing part. We have assessed our proposed method performance using six widely available benchmark test sequences and three echocardiography image sequences. The evaluations confirm that while our proposed method keeps the PSNR and SSIM performance suitable, it is at least twice faster in comparison with reference methods on the benchmark dataset. The echocardiography evaluations reveal our proposed method superiority in terms of PSNR, SSIM, and computation time, which make the proposed method suitable as an echocardiographic-specific method.

This paper proposes a parallel image thinning algorithm and a skeletonization algorithm based on cellular automaton (CA). Cellular automaton is a parallel computation model and a non-linear dynamical system. In this paper, each image pixel is identified as a cell of CA and the change of cell depends on the current state of itself and the state of its neighbors. In a binary image, this paper assumes that the objects (white pixel) are preys which are surrounded by many ants (every black pixels). The movement of ants is controlled by cellular automation. The ants gnaw preys until the preys (objects) become skeleton. The proposed parallel skeletonization algorithm can produce a traditional skeleton with a thin line located in the center of object, and the proposed thinning algorithm can produce a new kind of skeleton which is named as the ants-gnawing skeleton. The computation of ants-gnawing skeleton is faster than the traditional skeleton while it contains more the structural features of image. Benefiting from the properties of cellular automation, the proposed thinning algorithm does not change the basic geometry structure of image, and it is invariant for image rotation.

In this paper a parametric point spread function (PSF) estimation method is presented. This method can be employed for estimating the parameters of linear motion blur vector. The proposed method works on single image and estimates angle and length of motion blur vector to generate required PSF for deblurring. This method is based on step-by-step estimation of motion blur vector. In the first approximation the blind method considers short length vectors in all directions and deblurs image with PSF of these candidate vectors. The quality of deblurred image is assessed by a no-reference quality measurement metric which is proposed in this paper. The proposed no-reference image quality metric evaluates degradation in sharpness of edges and the amount of resulted artifact caused by saturated pixels, in the deblurred image. Those motion vectors that caused unacceptable deblurring results are omitted in the next iteration. The approximation is improved by increasing the length of the remaining vectors and the same process continues iteratively. The process goes on until only one vector remains as the estimation for motion blur vector. Experimental results show that in estimation of motion blur vectors, the length estimation error is less than one pixel in 85% of the cases and angle estimation error in 95% of the cases is less than one degree. Comparing with a conventional method indicates that the proposed method shows more than four times improvement in length estimation and 10% improvement in angle estimation. Moreover, the proposed method has comparatively lower computational load than other conventional deblurring methods.

In order to take advantage of the powerful advantages of heterogeneous devices and improve the robustness of the vehicle-mounted surround vision algorithm(VSVA), several key technologies of VSVA are improved in the paper. Firstly, computationally intensive tasks are calculated by heterogeneous Graphics Processing Unit(GPU), at the same time, so as to adapt to the VSVA, the memory model and computing model of GPU are optimized. Then a perspective transformation algorithm based on geometric constraints is proposed to improve the quality of the transformed image. Finally, an image alignment and fusion algorithm based on a calibration board is proposed, which reduces the complexity of the algorithm while ensuring the robustness of the image fusion algorithm. The paper compares the proposed algorithm with the traditional algorithm, the test results show that the proposed algorithm has good robustness and the overall performance of the VSVA is improved to 95.39%, the proposed algorithm can be widely used.

This paper proposes a design of 2-D chaotic Baker map for image encryption which utilizes three modes of operations: 1) the cipher block chaining (CBC) mode, 2) the cipher feedback (CFB) mode, and 3) the output feedback (OFB) mode. The proposed image cryptosystem is characterized by a short encryption time of scalevariant images and a high level of confusion and diffusion due to its shuffling and substitution processes. This is useful in applications such as online streaming of paid videos, in which both the speed of encryption\decryption and a good encryption quality is required. A comparison between the proposed image cryptosystem, the traditional 2-D chaotic Baker map permutation cryptosystem, and the RC6 substitution cryptosystem is presented in the paper. A comparison is held with relevant techniques and the results reveal that the proposed image cryptosystem achieves a high degree of security. It is also more immune to noise than the RC6 cryptosystem and takes less processing time for images with large dimensions than both the chaotic cryptosystem and the RC6 cryptosystem. The superiority of the proposed cryptosystem has been proved for image encryption against the recent techniques from the cryptographic viewpoint.

Rate-Distortion (RD) performance of Distributed Video Coding (DVC) is considerably less than that of conventional predictive video coding. In order to reduce the performance gap, many methods and techniques have been proposed to improve the coding efficiency of DVC with increased system complexity, especially techniques employed at the encoder such as encoder mode decisions, optimal quantization, hash methods etc., no doubt increase the complexity of the encoder. However, low complexity encoder is a widely desired feature of DVC. In order to improve the coding efficiency while maintaining low complexity encoder, this paper focuses on Distributed Residual Video Coding (DRVC) architecture and proposes a simple encoder scheme. The main contributions of this paper are as follows: 1) propose a bit plane block based method combined with bit plane re-arrangement to improve the dependency between source and Side Information (SI), and meanwhile, to reduce the amount of data to be channel encoded 2) present a simple iterative dead-zone quantizer with 3 levels in order to adjust quantization from coarse to fine. The simulation results show that the proposed scheme outperforms DISCOVER scheme for low to medium motion video sequences in terms of RD performance, and maintains a low complexity encoder at the same time.

Domain adaptation is widely-used in deep neural networks to address the problem of data distribution shift. Most of the deep CNN models use the Maximum Mean Discrepancy(MMD) to measure the distribution difference between the source and task domains, which have achieved great success on transfer learning tasks. However, these conventional domain adaptation methods have limited use in dynamic knowledge adaptation due to the constant transfer coefficient for all adaptation layers. In this paper, we propose an efficient dynamic domain adaptation method on deep CNN models, which transfers knowledge dynamically according to different layers and training extent. Specifically, we first present a deep understanding of how transferable each layer in various deep CNN models is, including the different VGG and AlexNet architectures. Then, a dynamic transfer coefficient is proposed based on the deep understanding. By doing this, our paper gives guidance on how to choose transferring layers and adaptation coefficients aptly instead of using empirical constant transfer parameters in conventional methods. Extensive experiments conducted on standard benchmark datasets demonstrate that the proposed method achieves the state-of-the-art results by using dynamic domain adaptation parameters compared with conventional methods.

In this paper we deal with the problem of calculating Automatic Exposure (AE) in digital cameras. The main problem that often occurs when taking pictures is correct exposure setting. Typically, smartphones with built-in cameras, as well as “cheap” compact digital cameras do not offer possibility of manual exposure setting. The reason is that users do not have knowledge how to set the optimal exposure, or just simply do not want to do this. Therefore, it forces that user has to rely on automatic exposure algorithms implemented in the camera. Unfortunately, these algorithms often do not perform well what causes improperly exposed images. In this paper, new algorithms for automatic exposure are proposed with the special focus on minimizing overexposed areas in the images. We have implemented proposed algorithms and conducted experiments for their efficiency, comparing with some modern cameras or smartphones. Experimental verification (enhanced by statistical analysis) shows that proposed algorithms give statistically less overexposed areas than comparative AEs.

Evolutionary algorithms have found enormous applications in solving real-world problems due to their stochastic nature. They have a set of control parameters, which are used to perform certain operations to induce randomness, scalar displacement etc. Various works have been done for tuning these parameters, as appropriate parameter tuning can enhance the performance of algorithm greatly. Inertia weights based parameter tuning is one of the widely used techniques for this purpose. In this paper, we have reviewed some of the inertia weight strategies and applied them to Social Group Optimization (SGO) to study the changes in its performance and have performed a thorough analysis on the same. Following the analysis, the need of a more generalized inertia weight strategy was felt which could be used in parameter tuning for different variety of problems and hence Sigmoid adaptive inertia weight have been proposed. SGO with sigmoid-adaptive inertia weight (SGOSAIW) has been simulated on twenty-seven benchmark functions suite and further simulated on few mechanical and chemical engineering problems and compared to other similar algorithms for performance analysis. In eight-benchmark function suite, SGOSAIW obtained better minima except one i.e. ‘Schwefel 2.26’ with respect to other algorithms investigated in this work. In nineteen-benchmark function suite, SGOSAIW obtained better minima except one i.e. ‘Noisy function’. Thus, the proposed algorithm yielded promising results which are well represented with suitable tables and graphs in the paper.

In recent years, Speech Emotion Recognition (SER) has received considerable attention in affective computing field. In this paper, an improved system for SER is proposed. In the feature extraction step, a hybrid high-dimensional rich feature vector is extracted from both speech signal and glottal-waveform signal using techniques such as MFCC, PLPC, and MVDR. The prosodic features derived from fundamental frequency (f0) contour are also added to this feature vector. The proposed system is based on a holistic approach that employs a modified quantum-behaved particle swarm optimization (QPSO) algorithm (called pQPSO) to estimate both the optimal projection matrix for feature-vector dimension reduction and Gaussian Mixture Model (GMM) classifier parameters. Since the problem parameters are in a limited range and the standard QPSO algorithm performs a search in an infinite range, in this paper, the QPSO is modified in such a way that it uses a truncated probability distribution and makes the search more efficient. The system works in real-time and is evaluated on three standard emotional speech databases Berlin database of emotional speech (EMO-DB), Surrey Audio-Visual Expressed Emotion (SAVEE) and Interactive Emotional Dyadic Motion Capture (IEMOCAP). The proposed method improves the accuracy of the SER system compared to classical methods such as FA, PCA, PPCA, LDA, standard QPSO, wQPSO, and deep neural network, and also outperforms many state-of-the-art recent approaches that use the same datasets.

Henon chaotic system is a kind of classical chaotic system which is widely studied and applied. However, when we analyze its chaotic characteristics, we can still find some shortcomings. In this paper, a Henon chaotic system with nonlinear terms is proposed and applied to image encryption. Our scheme first modifies the linear term y(n) of the original Henon chaotic system to the nonlinear term ey(n), and then obtains the new Henon chaotic system. Then, a series of methods such as bifurcation graph, Lyapunov exponent, linear complexity, histogram, balance, run characteristics and NIST test were used to verify the performance improvement of the new Henon chaotic system. And then, the random sequences generated by the new Henon chaotic system are combined with the image encryption algorithm to complete the image encryption through three steps: global confusion, bit recombination and pixel value diffusion. Finally, the security of the encryption algorithm is verified by secret key space analysis, anti-statistical attack analysis, anti-differential attack analysis, anti-shearing and noise attack analysis. In this paper, the improvement of Henon chaotic system and the design of encryption algorithm all have achieved good results.

This paper proposes an encryption algorithm that uses the initial values and parameters of the chaotic system as the key, and mainly uses the similar deoxyribonucleic acid (DNA-like) coding method and the similar Zigzag (Zigzag-like) transform to encrypt the image. Firstly, the image is pre-scrambled by the method of Zigzag-like transformation, and then the second scrambling is performed by a sorting scrambling algorithm with identification value. Secondly, the image is diffused by DNA-like coding method. Finally, the image is again diffused using the ortho exclusive OR (XOR) method with chaotic perturbation terms. The experimental results show that the chaotic image encryption algorithm proposed in this paper has satisfactory results. In addition, the algorithm is compared to the previously proposed chaotic image encryption algorithm for the Zigzag transform method or the deoxyribonucleic acid (DNA) coding method. The contribution is to improve the Zigzag transformation method and the DNA coding method, so that it has the advantages of higher security and higher sensitivity. It can also effectively resist exhaustive and differential attacks with better statistical characteristics.

This paper presents an improved image-adaptive watermarking technique. Two image watermarks are embedded in the high entropy 8 × 8 blocks of the host image. DWT is applied on these blocks using the principle of sub band coding. This decomposes the high entropy blocks into four sub band coefficients, wherein the approximation and vertical frequency coefficients are modeled using Gaussian (or Normal) distribution. The two watermarks are inserted in the host image using Adjustable Strength Factor (ASF). It is calculated adaptively using the fourth statistical moment known as kurtosis. A limited side information is also transmitted along with the watermarked image. This side information consists of high entropy block positions and Gaussian distribution parameters. To extract both watermarks from the received watermarked image, the high entropy block positions sent in the side information help in applying DWT to calculate the approximation and vertical frequency coefficients. Gaussian (or Normal) distribution is similarly used for modeling and calculating the distribution parameters. This helps the Maximum Likelihood (ML) decoder to recover the watermarks successfully using a statistical approach. Two important contributions are presented in this paper. Firstly, adjustable kurtosis values are used which improves the capacity and robustness of the proposed technique. Secondly, the proposed work is implemented on medical applications and gives better performance as compared to the existing methods. Further, the efficiency of the proposed work is evaluated by better simulation results using PSNR, NCC, SSIM and GMSD under different attacks. The technique is highly robust as watermarks survive under different attacks. This increases security and ensures copyright protection.

The main objective of the image enhancement is to improve the visual appearance or quality of an image. In this paper, the proposed scheme aims to improve the performance of the homomorphic filtering by employing the fractional derivatives with Discrete Fourier Transform (DFT) and Fractional Fourier Transform (FrFT). FrFT in combination with fractional derivative provides two fractional orders as extra degrees of freedom, thus, providing more design flexibility. This paper uses Grunwald-Letnikov (GL) fractional derivative to enhance the high and mid frequency components non-linearly while preserving the low frequency components. In the proposed approach, modification of homomorphic filtering technique is done on the basis of fractional derivative and FrFT to enhance the low contrast and non-uniformly illuminated images. The effectiveness of the proposed work is evaluated on the basis of various image assessment parameters such as PSNR, information entropy, universal image quality index, etc. on several images of different sizes. The proposed scheme outperforms the existing state-of-the-art techniques by providing better image visual quality and image information in terms of average PSNR and entropy values. The improvement in the average PSNR and information entropy is in the range 0.2635–50.37 dB and 0.02–42% respectively for standard images as well as for images with different contrast and illumination conditions.

This paper presents a novel reversible data hiding into a Vector Quantization (VQ) and Side Match Vector Quantization (SMVQ) based compression image to embed high capacity secret bits and recover cover image after data extraction. For optimal embedding capacity and to achieve exact recovery of cover image, this paper uses Enhanced Imperialist Competitive Algorithm (EICA). The threshold value is determined by the fitness function contrast sensitivity in EICA in order to signify embedding rate of each region in a cover image based on the size of the secret message. During data hiding, the output size of code stream is preserved in hiding two secret bits into a single index value. Discrete Cosine Transform (DCT) and Burrows Wheeler Transform (BWT) is applied before quantization for exact recovery of cover image and to achieve high compression ratio. Excellent energy compaction is provided by DCT and BWT reorders the symbols according to their context. Thus the proposed method provides a novel technique to embed secret bits into the cover image and compresses the embedded image. The output will be in the form of code streams with preserved size. The experimental results show that the proposed technique achieves high embedding capacity and compression rate.

The paper mainly analyzes four typical reversible data hiding (RDH) algorithms that use PEE technology in combination: Conventional Prediction-error-expansion(C-PEE), Adaptive Prediction-error-expansion(A-PEE), Pairwise Prediction-error-expansion (P-PEE), Improve Pairwise Prediction-error-expansion(IP-PEE). And the experimental comparison results are also given. In addition, in the research process, we found that the peak signal-to-noise ratio (PSNR) and embedding capacity (EC) of many algorithms are very close, due to lack of benchmark tools for technical evaluation in the RDH field, thus this paper proposes a new performance measurement method based on the PSNR and EC measurement, we call it the prediction error accuracy rate, Determined by the ratio of the capacity that meets the embedding conditions to the total capacity of the image, combined with the measurement of this method, can better determine whether the performance of the algorithm is optimal. Finally, in this area, challenge and future directions are also given.

In this paper, a novel image encryption algorithm based on a new discrete chaotic system is presented. The chaotic characteristics of the new chaotic system are analyzed by phase diagram, Lyapunov exponent, bifurcation diagram and complexity analysis. The randomness of chaotic sequences is test by the NIST SP 800-22 test package. The calculation time with different length sequence is recorded. Based on the analyses results, a flexible encryption scheme is designed for 2D image and 3D image. The hash value which is calculated from the SHA-256 hash function is used to change the initial conditions of the discrete system firstly. And then, the chaotic sequences are used to scramble and spread the value for the images by Arnold matrix and DNA diffusion algorithm. The security analysis shows that the proposed encryption algorithm possesses higher security features to preserve the subject and resist conventional attack. The results of this paper offer a different realization scheme for protection of image files.

Compressive sensing is widely used to image compression and encryption algorithms due to its high efficiency, but the existing algorithms have some flaws and insufficiency such as low reconstruction quality, small key space and weak security. Therefore, in this paper, a novel 5D chaotic system is proposed, which has larger key space and more complex key stream. According to the proposed 5D chaotic system, an image compression and encryption algorithm based on compressive sensing and nonlinear diffusion is proposed. In addition, in order to improve the image reconstruction quality of compressive sensing, an algorithm is proposed in this paper to optimize the measurement matrix of compressive sensing. Theoretical analysis shows that the proposed 5D chaotic system is chaotic and it shows many superior properties. The algorithm proposed to optimize the measurement matrix is also proved effective for reconstruction quality. The simulation results show that our algorithm has advantages in compression performance, key sensitivity, key space and time complexity, and it can also resist statistical attack and other common attacks.

Super-resolution reconstruction plays an important role in reconstructing image detail and improving image visual effects, which is achieved mostly by linear interpolation algorithm in the existing display equipment, and there are no obvious details and blurred edges. In this paper, a multi-connection convolution network is proposed to solve the problems of the super-resolution reconstruction (SR) algorithm. The network constructs a multi-connection network structure, where a long skip strategy is used to obtain the identity mapping. It can concatenate the low-level features and high-level features, and can simultaneously represent various complex reconstruction scenes. In addition, a strong and flexible two parameter loss function is used to optimize and train the deep network and improve the generalization ability of the network model. The simulation results show that the proposed SR algorithm in this paper can generate high resolution images with rich details and clearness, and the objective quantitative evaluation is greatly improved.

With the development of computer technology and network technology, a large amount of printed documents are converted to the electronic documents and spread on internet. When a halftone image is scanned to an electronic document, the screen patterns will appear, so the inverse halftoning algorithms are needed to remove the screen patterns and improve image quality. In this paper, the halftoning techniques are introduced firstly, then this paper reviews different inverse halftoning algorithms. The inverse halftoning algorithms introduced in this paper include the low-pass filter algorithm, the fast algorithm, the wavelet based algorithm, the maximum posteriori probability algorithm, the LUT algorithm, the vector based algorithm, and the deconvolution based inverse halftoning algorithm. The image quality evaluation of these inverse halftoning algorithms is also discussed. Finally this paper summarizes the shortcomings of current inverse halftoning algorithms and the directions that can be improved in the future.

Due to its high query speed and low storage cost, binary hashing has been widely used in approximate nearest neighbors (ANN) search. However, the binary bits are generally considered to be equal, which causes data points with different codes to share the same Hamming distance to the query sample. To solve the above distance measure ambiguity, bitwise weights methods were proposed. Unfortunately, in most of the existing methods, the bitwise weights and the binary codes are learnt separately in two stages, and their performances cannot be further improved. In this paper, to effectively address the above issues, we propose an adaptive mechanism that jointly generate the bitwise weights and the binary codes by preserving different types of similarity relationship. As a result, the binary codes are utilized to obtain the initial retrieval results, and they are further re-ranked by the weighted Hamming distance. This ANN search mechanism is termed AR-Rank in this paper. First, this joint mechanism allows the bitwise weights and the binary codes to be used as mutual feedback during the training stage, and they are well adapted to one other when the algorithm converges. Furthermore, the bitwise weights are required to preserve the relative similarity which is consistent with the nature of ANN search task. Thus, the data points can be accurately re-sorted based on the weighted Hamming distances. Evaluations on three datasets demonstrate that the proposed AR-Rank retrieval system outperforms nine state-of-the-art methods.

Side match vector quantization (SMVQ) is a widely used image compression algorithm for data hiding applications. Compared with conventional vector quantization (VQ) algorithm, a smaller and more powerful state codebook (SC) which is generated by utilizing the correlation in natural image is used in SMVQ to achieve low bit rate. However, the visual quality of reconstructed image by using SMVQ is significantly decreased. In this paper, a novel low bit rate coding algorithm named structured SMVQ (SSMVQ) is proposed. The size of SSMVQ’s SC is flexible and the SC of SSMVQ is composed by a smaller SC of conventional SMVQ and a supporting codebook which is newly introduced in this paper. Experimental results show that the proposed structed SMVQ is able to achieve satisfactory PSNR when the bit rate is extremely low.

Target tracking is one of the pivotal technologies in intelligent video surveillance systems. Facing the complex and various scenarios in practical applications, improving the accuracy and real-time of target detection and tracking is has become the goal of current monitoring systems. Firstly, the target feature expression model is established by fusing Sobel Median Binary Pattern (SMBP) and H-S features while the final target probability model is set up by a weighted color kernel function histogram. Secondly, the final target probability model is established by fusing a weighted color kernel function histogram. Thirdly, the improved unscented Kalman particle filtering algorithm proposed in this paper is embedded in the target tracking framework to complete the target tracking. Lastly, compared with the traditional tracking algorithm, the experiments results show that the target tracking algorithm proposed in this paper improves the tracking accuracy by about 4%.

In remote sensing image fusion field, traditional algorithms based on the human-made fusion rules are severely sensitive to the source images. In this paper, we proposed an image fusion algorithm using convolutional neural networks (FusionCNN). The fusion model implicitly represents a fusion rule whose inputs are a pair of source images and the output is a fused image with end-to-end property. As no datasets can be used to train FusionCNN in remote sensing field, we constructed a new dataset from a natural image set to approximate MS and Pan images. In order to obtain higher fusion quality, low frequency information of MS is used to enhance the Pan image in the pre-processing step. The method proposed in this paper overcomes the shortcomings of the traditional fusion methods in which the fusion rules are artificially formulated, because it learns an adaptive strong robust fusion function through a large amount of training data. In this paper, Landsat and Quickbird satellite data are used to verify the effectiveness of the proposed method. Experimental results show that the proposed fusion algorithm is superior to the comparative algorithms in terms of both subjective and objective evaluation.

The existed digital steganography models and theories are not effective enough to guide the steganography processing. Based on previous studies, this paper proposes a complete digital steganography model based on additive noise. And then, with security analysis from KL divergence, the embedding optimization strategy is given through theoretical derivation needless of any side information: optimizing the embedding modification position and optimizing the embedding modification direction (+1 or − 1). Through theoretical derivation, we also obtain the quantitative relationship between the pixels modification probability and the adjacent pixels difference, and prove that modification by ±1 randomly cannot enhance steganographic security definitely. The research in this paper can provide theoretical guidance for the design of steganography algorithms. Compared with previous studies, the proposed embedding optimization strategy has outstanding advantages of being easy to implement and being effective to improve steganographic security. The experiments by optimizing LSBM and MG algorithms show that the proposed embedding optimization strategy can effectively improve each algorithm’s steganographic security at a relative small payload.

Multimedia communication is revolutionizing all major spheres of human life. The advent of IoT and its applications in many fields like sensing, healthcare and industry, result exponential increase in multimedia data, that needs to be shared over insecure networks. IoT driven setups are however constrained in terms of resources as a result of their small size. From data security point of view a conventional algorithms cannot be used for data encryption on an IoT platform given the resource constraints. The work presented in this paper studies the performance of SIMON cryptographic algorithm and proposes a light-weight-cryptography algorithm based on SIMON for its possible use in an IoT driven setup. The focus is on speed enhancement benefitting from software prospective, making it different than common studies mostly reflecting hardware implementations. To achieve performance in practical prospective, the contribution looks into SIMON cipher’s characteristics considering utilizing it for internet of things (IoT) healthcare applications. The paper suggests further improvement to implement the original SIMON cryptography in order to reduce the encryption time and maintain the practical trade-off between security and performance. The proposed work has been compared to Advanced Encryption Standard (AES) and the original SIMON block cipher algorithms in terms of execution time, memory consumption. The results show that the proposed work is suitable for securing data in an IoT driven setup.

Automatic segmentation of land use and land cover from high resolution remote sensing imagery has been an essential research area in image processing for the past two decades. Timely and reliable information of land use and land cover is very much essential in administration for proper planning and decision making in various areas like agriculture, urban development, environment protection, etc. In this paper, a new algorithm ULCDSA (Unsupervised Linear Contact Distributions Segmentation Algorithm) is proposed for unsupervised segmentation of high resolution panchromatic data. Texture features extracted with the help of linear contact distributions and mathematical morphology are used in this paper. The proposed method has been implemented and tested on various panchromatic images of N.R.S.C’s Cartosat-II data sets, Google earth images and on other aerial images. The results have then been compared with gray scale co-occurrence matrix algorithms and promising results have been obtained.

Unsharp Masking (UM) is a popular technique widely used for sharpening medical imagery. However, UM has two operational parameters which have crucial influence on its performance. They are amount (λ) and threshold (α). Improper selection of the threshold and amount will cause noise amplification and overshoot artefact, respectively. A fully adaptive UM with data driven operational parameters, for sharpening Magnetic Resonance (MR) images is proposed in this paper. The proposed sharpening scheme is compared with the homomorphic filter in terms of sharpness of the output image, width of salient edges, feature preservation, saturation and edge quality degradation due to noise. The proposed scheme of UM is found to be superior to homomorphic filter in terms of edge strength and feature preservation and it is free from overshoot artefacts. Modified configuration of UM, proposed in this paper can be used for improving the edge quality of MR images. It can be incorporated as a plug-in in software tools used for the automated analysis of MRI.

This paper presents a new method for BEMD. BEMD can decompose a source image into several two-dimensional intrinsic mode functions. During the image decomposition process, it is required to interpolate and draw the upper and lower envelopes. However, these interpolations and drawing the enveloping surface require a large amount of computing time and artificial screening. Thus, some scholars proposed the rapid realization of BEMD. Further, the window size being fixed during the decomposition process led to losses in the data-driven characteristics, adaptability and dimension consistency of the original BEMD. Therefore, this paper proposes a simple but effective means of keeping the original BEMD method features. The estimate reconstruction method is used to replace surface interpolation, and the variable neighborhood window method is adopted to replace the fixed neighborhood window method. In this article, an order filter is used to reconstruct the upper and lower envelopes, and then the filter size is obtained by using the fact that the image information itself is adaptively variable. Through an empirical analysis, this paper shows that this method can keep the original BEMD method’s rapid decomposition, data-driven characteristics, adaptivity and consistency of scale.

In this paper, a multi-objective bird swarm algorithm (MOBSA) is proposed to cope with multi-objective optimization problems. The algorithm is explored based on BSA which is an evolutionary algorithm suitable for single objective optimization. In this paper, non-dominated sorting approach is used to distinguish optimal solutions and parallel coordinates is applied to evaluate the distribution density of non-dominated solution and further update the external archive when it is full to overflowing, which ensure faster convergence and more widespread of Pareto front. Then, the MOBSA is adopted to optimize benchmark problems. The results demonstrate that MOBSA gets better performance compared with NSGA-II and MOPSO. Since a vehicle power train problem could be treated as a typical multi-objective optimization problem with constraints, with integration of constrained non-dominated solution, MOBSA is adopted to acquire optimal gear ratios and optimize vehicle power train. The results compared with other popular algorithm prove the proposed algorithm is more suitable for constrained multi-objective optimization problem in engineering field.

With the emergence of Wireless multimedia transmission system, the distribution of multimedia contents has now become a reality. To solve the problem of stability in the process of transmission, this paper proposes an improved channel estimation with Bayesian framework based on compressed sensing algorithm in multimedia transmission system. The algorithm uses the sparse characteristics of the channel and can reduce the pilot sequence length under the same conditions. Due to the high complexity of the support agnostic Bayesian matching pursuit algorithm, our algorithm improves the support set, which proposed Expectation Prune Matching Pursuit algorithm in the paper. At each sparsity level of the channel, an expanded support set is given by adding some positions corresponding to the atoms that have a larger inner product value with the current residual signal. Then the best support set is obtained by removing the wrong positions and adopting the idea of Bayesian estimation algorithm in the expanded support set. The estimated channel and the relative probability of the best support set at each sparse level are calculated. Finally, the expectation of the channel is calculated and regarded as the estimation of the channel. Compared with comparison algorithm in the error and bit error rate under different SNR conditions, our proposed algorithm can reduce the computational complexity while maintaining the estimation accuracy.

In this paper, a new iterative algorithm has been presented by aggregating Stationary Wavelet Transform (SWT), Bilateral filtering, Bayesian estimation, and Anisotropic Diffusion (AD) filtering to reduce the speckle noise in SAR images. For this purpose, speckle images were first decomposed using two-dimensional stationary wavelet transform and then a suitable filtering method was used to filter respective coefficients of each sub-band of the speckled images. Generally, in wavelet transform-based noise reduction methods, filtering and thresholding techniques are usually applied to the coefficients of the detail sub-bands and the residual speckle noise is ignored in the approximate sub-band. In this paper, bilateral filtering has been applied to reduce the speckle noise in the approximate sub-band. We used Bayesian estimator to calculate the noise-free signal in the horizontal and vertical sub-bands with respect to that some parts of signal coefficients are eliminated in the traditional thresholding techniques. Moreover, we applied anisotropic diffusion filtering method to preserve the edges and structure of image along the diagonal subband which has more details (the entropy is maximum) than other directions in radar and optic images. Finally, both the proposed algorithm and other speckle noise reduction methods were applied on two synthetic speckled images and an actual SAR image in San Francisco. Their efficiencies were compared according to the Structural SIMilarity(SSIM), Peak Signal to Noise Ratio (PSNR), Equivalent Number of Looks (ENL), Speckle Suppression Index (SSI) and Speckle Suppression and Mean Preservation Index (SMPI). The experimental results indicate that the proposed algorithm efficiently reduces the speckle noise and preserves the edges and structure of image.

This paper proposes a selective encryption (SE) algorithm for the almost recently video compressing technique that is called High-Efficiency Video Coding (HEVC). The improvement of our HEVC SE algorithm is to secure real-time HEVC streaming using low computational overhead, same bit rate and ensuring the video format compliance. This contribution is achieved with low computational RC6 algorithm to encrypt subset of binstrings that is binarized using the bypass binary arithmetic coding mode in the context adaptive binary arithmetic coding (CABAC) process of HEVC. This binstrings subset is the non-zero discrete cosine transform (DCT) coefficients sign bits, motion vector difference (MVD) sign bits, remaining absolute values suffix of DCT, MVD absolute values suffix, Sample adaptive offset (SAO) sign bit, residual size, reference picture index and the delta Quantization Parameter (QP). Also, this paper investigates the performance evaluation of using the RC6 with its operation modes within the proposed HEVC CABAC SE. This investigation is done for choosing the best operation mode for RC6 to be suitable for being used in the HEVC real-time application. The security analysis like histogram analysis, correlation coefficients test and key sensitivity test are presented to make sure the protection of HEVC CABAC SE algorithm against brute force and statistical attacks. The performance analysis results prove that the HEVC CABAC SE is highly secure and can be used in real-time HEVC applications.

Digital images are widely distributed today over the internet and through other mediums. There are powerful digital image processing tools which have made perfect image duplication a trivial procedure. Therefore, security, copyright protection, and integrity verification of digital video have become an urgent issue in the digital world, which can be achieved by means of a technique called digital watermarking. The issues of watermarking are to achieve imperceptibility, robustness, payload and security simultaneously. This paper presents a new Quaternion Curvelet Transform (QCT) based video watermarking technique for embedding a video on another video in a secure and optimized way. However, many of the existing techniques are unable to handle the problem of rotation, translation and scaling attacks on watermarked video. This study presents a novel embedding approach where in the quaternion transform followed by the traditional Curvelet transform is able to overcome the above disadvantages. In this paper, we first represent the color cover video frames in the quaternion form and subsequently generate Quaternion Curvelet Transform (QCT) coefficients for each such quaternion frames. Second, in order to withstand the desynchronization causes due to the geometrical attacks, the Harris corner detection algorithm is used to determine the invariant feature points on the QCT coefficients, which is followed by the calculation of energy for each block centered on those feature points. Third, in order to maintain the trade-off among the watermarking characteristics, the optimized location for embedding the watermark frames is determined using the cuckoo search optimization algorithm. Finally, to attain the authenticity, an authentication key generation procedure is employed using the owner’s biometric image and the watermarked frames. The experimental results prove that the proposed method is promising in terms of robustness, imperceptibility and security to most notable image processing attacks, geometrical attacks, and video processing attacks among the various conventional watermarking algorithms.

In this paper, we propose an efficient transform-based blind audio watermarking technique by introducing a parametric quantization index modulation (QIM). Theoretical expressions for the signal to watermark ratio and probability of error are derived and then used in an optimization technique based on the Lagrange multipliers method to find the optimal values for the parameters of the parametric QIM that ensure the imperceptibility while maximizing the robustness under an additive white Gaussian noise (AWGN) attack. Moreover, a fast scheme for the implementation of the proposed watermarking technique is developed and an efficient procedure is suggested to find the interval for the best selection of the watermark embedding positions that provide a good trade-off between the effects of high and low pass filtering attacks. The parameters of the resulting optimal parametric QIM coupled with the embedding positions constitute a highly robust secret key for the proposed watermarking technique. We also carry out several experiments to show the usefulness of the theoretical analysis presented in the paper and compare the proposed technique with other existing QIM-based watermarking techniques by considering known attacks such as AWGN, re-quantization, resampling, low/high pass filtering, amplitude scaling and common lossy compressions.

This paper joins the image-matching module into the hybrid recommendation system and constructs its workflow, which fuses image features and hybrid recommendation algorithms to improve diversity of advice result. SIFT feature extracted was used as the standard of image matching, improved LSH algorithm based on p-stable distribution to implement image matching and searching module for high dimensional and huge image set, then redesigned the workflow of existing commodity recommendation system combined with the proposed image matching module. This paper proposed an improved LSH algorithm based on p-stable distribution to finish image searching and matching. The experiment proved that the algorithm has a certain degree of optimization to improve recall rate and error rate at the same time, through the matching time and the length of hash table shows that the algorithm optimizes the memory utilization and search efficiency. The extracted SIFT feature of images is the only foundation we used when comparing different images at present. In the following research, we can try to use a variety of image features as the basis for matching to improve the reliability of the matching results.