We present a nonconstraining and low-cost online signature acquisition system that has been developed to enhance the performances of an existing multimodal biometric authentication system (based initially on both voice and image modalities). A laboratory prototype has been developed and validated for an online signature acquisition.

We present a practical hexagonal storage and addressing scheme, which eliminates the difference between theory and implementation in other addressing methods. This scheme employs a middleware-based address-mapping module that separates the algorithm and specific data addressing; thus any hexagonal algorithm can keep the native and consistent forms as in the coordinate system through theory and implementation. The scheme simplifies the implementation work and preserves all excellent features of the hexagonal lattice. Finally, we discuss the implementation issues and show that it’s feasible and can be implemented efficiently.

For many photographers today, mobile photography is photography. According to statistics available on the photo-sharing site CrossRef

XSLOpenURL/

, over the last year the most popular camera used to take photos on the site has been an Apple iPhone. In fact, during the summer of 2012, the iPhone 4 and the iPhone 4S occupied the number one and number two spots. The success of mobile photography perhaps goes hand in glove with the rise of social networking sites. Mobile photography has enabled any individual with a cell phone (the ubiquitous mobile camera) to quickly and easily share visual images of their lives. Mobile cameras are cheap, small, lightweight, and connected—pictures can be taken anytime and anywhere—and shared immediately.

The scaling limits of multi-aperture systems have been widely discussed from an information-theoretical standpoint. While these arguments are valid as an upper limit, the real-world performance of systems for mobile devices remains restricted by optical aberrations. We argue that aberrations can be more easily controlled with certain architectures of multi-aperture systems, especially those manufactured on wafer scale (wafer-level optics, WLO). We complement our analysis with measurements of one single- and one multi-aperture WLO camera. We examine both sharpness and sensitivity, giving measurements of modulation transfer function and temporal noise, and showing that multi-aperture systems can indeed reduce size without compromising performance.

Computing depth maps from a stereo pair is a well-studied problem of computer vision, and a large number of methods and cost functions have been proposed. The methods have different strengths and weaknesses and different error characteristics. That is, a pixel could be assigned an erroneous depth value by some methods, but other methods could assign the correct depth value to the same pixel. We describe a method that can make use of multiple depth estimates of low quality and fuse them, by trying to retain the correct depth values and rejecting the incorrect depth values, in order to obtain a more accurate result. We observe that depth values of pixels located in smooth areas of a depth estimate and depth values which survive left-right cross validation tend to be more accurate. Our method makes use of a reliability criterion-based upon the smoothness and cross-validation of the depth estimates that allows us to patch the estimates together and obtain a higher quality result.

Long-exposure handheld photography is degraded with blur, which is difficult to remove without prior information about the camera motion. In this work, we utilize inertial sensors (accelerometers and gyroscopes) in modern smartphones to detect exact motion trajectory of the smartphone camera during exposure and remove blur from the resulting photography based on the recorded motion data. The whole system is implemented on the Android platform and embedded in the smartphone device, resulting in a close-to-real-time deblurring algorithm. The performance of the proposed system is demonstrated in real-life scenarios.

This paper concerns the compensation of specular highlight for handheld image projectors. By employing a projector-camera configuration, where the camera is aligned with the viewer, the distortion caused by nonideal (e.g., colored, reflective) projection surfaces can be estimated from the captured image and compensated for accordingly to improve the projection quality. This works fine when the viewing direction relative to the system is fixed. However, the compensation becomes inaccurate when this condition changes, because the position of the specular highlight changes as well. We propose a novel method that, without moving the camera, can estimate the specular highlight seen from any position and integrate it with Grossberg’s radiometric compensation framework to demonstrate how view-dependent compensation can be achieved. Extensive results, both objective and subjective, are provided to demonstrate the performance of the proposed algorithm.

Real-time image and video processing applications require skilled architects, and recent trends in the hardware platform make the design and implementation of these applications increasingly complex. Many frameworks and libraries have been proposed or commercialized to simplify the design and tuning of real-time image processing applications. However, they tend to lack flexibility, because they are normally oriented toward particular types of applications, or they impose specific data processing models such as the pipeline. Other issues include large memory footprints, difficulty for reuse, and inefficient execution on multicore processors. We present a novel software architecture for time-constrained machine vision applications that addresses these issues. The architecture is divided into three layers. The platform abstraction layer provides a high-level application programming interface for the rest of the architecture. The messaging layer provides a message-passing interface based on a dynamic publish/subscribe pattern. A topic-based filtering in which messages are published to topics is used to route the messages from the publishers to the subscribers interested in a particular type of message. The application layer provides a repository for reusable application modules designed for machine vision applications. These modules, which include acquisition, visualization, communication, user interface, and data processing, take advantage of the power of well-known libraries such as OpenCV, Intel IPP, or CUDA. Finally, the proposed architecture is applied to a real machine vision application: a jam detector for steel pickling lines.

A new automatic feature-based registration algorithm of multisensor images with affine deformation is presented. Although the feature-based registration methods have an advantage in reducing computational load over the area-based ones, certain typical issues such as complex deformation and grayscale discrepancy existing in a multisensor image pair will make the design of a robust feature descriptor challenging. To deal with these issues, in contrast with most existing feature-based methods that describe feature points directly, we introduce an additional procedure of feature quadrilateral construction in the proposed algorithm. Then the descriptors for the constructed feature quadrilaterals are designed based on the affine-invariance property of triangle area representation. By doing these, the proposed algorithm’s robustness to both affine deformation and grayscale discrepancy can be guaranteed. Besides, since the calculation of feature descriptors only involves simple algebraic operations, the proposed method has low computational load. Experimental results using real multisensor image pairs are presented to show the merits of the proposed method.

The mission of speckle filtering is more intricate for a polarimetric synthetic aperture radar (SAR) system than for a single polarization SAR system. A new speckle filter that employs nonlinear anisotropic diffusion is proposed. The speckle filtering principles in a polarimetric SAR system are thoroughly investigated. Our study demonstrates that preservation of polarimetric properties and reduction of speckle level are very important clauses among those principles. Anisotropic diffusion outperforms window-based filters because it utilizes a diffusion tensor that can bias the direction of diffusion toward the orientation of interesting features. The procedure of diffusion is iterative, so it can lead to a gradual alleviation of speckle noise. The parameters for diffusion are flexible and adjustable, thus the filtering results are controllable according to the requirements of specific applications. Multidimensional and vector-valued model is combined with local statistics based on polarimetric SAR configuration. The experimental results on electromagnetics institute synthetic aperture radar (EMISAR) datasets show that the proposed filter has good performance in speckle reduction and preservation of polarimetric properties.

An equidistant fish-eye camera can be calibrated using a single image of two sets of parallel lines in the scene, which is very convenient for practical applications. The core of the calibration procedure is the fitting of center collinear circles that intersect at two vanishing points. A fast and accurate fitting method that can be used for equidistant fish-eye camera calibration is presented. First, the fitting problem is formulated as a nonlinear least square problem and solved using Levenberg-Marquardt optimization. The objective function is also derived, as well as the Jacobian required by the Levenberg-Marquardt algorithm. It is also demonstrated how to determine the initial estimations. Experimental results on synthetic data have demonstrated the superiority of our method to two existing approaches in terms of speed and accuracy. Results on real data have also demonstrated its effectiveness for fish-eye camera calibration.

We improved an image segmentation algorithm based on neutrosophic set (NS) and extended the modified method into color image segmentation. The original NS image segmentation approach transformed the images into NS domain, which is described using three membership sets: T, I, and F. Then two operations, α-mean and β-enhancement operations were employed to reduce the set indeterminacy. Although this method was quite successful in image segmentation application, some drawbacks still exist, such as oversegmentation and fixed α and β parameters. Thus, a new algorithm is proposed to overcome these limitations of the NS-based image segmentation algorithm. Then, the new modified method is extended into color image segmentation. The NS-based image segmentation algorithm is applied to each color channel independently. Then each channel is moved to a matrix column, respectively, to construct the input matrix to the γ-K-means clustering. Experiments are conducted on a variety of images, and our results are compared with those new existing segmentation algorithm. The experimental results demonstrate that the proposed approach can segment the color images automatically and effectively.

Image restoration and deconvolution from blurry and noisy observation is known to be ill-posed. To stabilize the recovery, total variation (TV) regularization is often utilized for its beneficial edge in preserving the image’s property. We take a different approach of TV regularization for image restoration. We first recover horizontal and vertical differences of images individually through some successful deconvolution algorithms. We restore horizontal and vertical difference images separately so that each is more sparse or compressible than the corresponding original image with a TV measure. Then we develop a novel deconvolution method that recovers the horizontal and vertical gradients, respectively, and then estimate the original image from these gradients. Various experiments that compare the effectiveness of the proposed method against the traditional TV methods are presented. Experimental results are provided to show the improved performance of our method for deconvolution problems.

A novel video resizing algorithm that preserves the dominant contents of video frames is proposed. Because a visual correlation (a similarity) exists on consecutive frames within an identical shot, the energy distribution of the neighboring frames is also correlated and similar, and then the seams in a frame are analogous to those of neighboring frames. Thus, the seams of the current frame are derived from a specified range considering the coordinates of the seams of the previous frame. The proposed method determines the two-dimensional connected paths for each frame by considering both the spatial and temporal correlations between frames to prevent jittering and jerkiness. For this purpose, a new temporal forward energy is proposed as the energy cost of a pixel. The proposed algorithm has a fast processing speed similar to the bilinear method, while preserving the main content of an image to the greatest extent possible. In addition, because the memory usage is remarkably small compared with the existing seam carving method, the proposed algorithm is usable in mobile devices, which have many memory restrictions. Computer simulation results indicate that the proposed technique provides a better objective performance, subjective image quality, and content conservation than conventional algorithms.

A (t,n) secret image-sharing scheme shares a secret image to n participants, and the t users recover the image. During the recovery procedure of a conventional secret image-sharing scheme, cheaters may use counterfeit secret keys or modified shared images to cheat other users’ secret keys and shared images. A cheated secret key or shared image leads to an incorrect secret image. Unfortunately, the cheater cannot be identified. We present an exponent and modulus-based scheme to provide a tamper-proof secret image-sharing scheme for identifying cheaters on secret keys or shared images. The proposed scheme allows users to securely select their secret key. This assignment can be performed over networks. Modulus results of each shared image is calculated to recognize cheaters of a shared image. Experimental results indicate that the proposed scheme is excellent at identifying cheated secret keys and shared images.

Number plate (NP) binarization and adjustment are important preprocessing stages in automatic number plate recognition (ANPR) systems and are used to link the number plate localization (NPL) and character segmentation stages. Successfully linking these two stages will improve the performance of the entire ANPR system. We present two optimized low-complexity NP binarization and adjustment algorithms. Efficient area/speed architectures based on the proposed algorithms are also presented and have been successfully implemented and tested using the Mentor Graphics RC240 FPGA development board, which together require only 9% of the available on-chip resources of a Virtex-4 FPGA, run with a maximum frequency of 95.8 MHz and are capable of processing one image in 0.07 to 0.17 ms.

Left ventricular (LV) torsion is a sensitive and global index of LV systolic and diastolic function, but how to noninvasively measure it is challenging. Two-dimensional echocardiography and the block-matching based speckle tracking method were used to measure LV torsion. Main advantages of the proposed method over the previous ones are summarized as follows: (1) The method is automatic, except for manually selecting some endocardium points on the end-diastolic frame in initialization step. (2) The diamond search strategy is applied, with a spatial smoothness constraint introduced into the sum of absolute differences matching criterion; and the reference frame during the search is determined adaptively. (3) The method is capable of removing abnormal measurement data automatically. The proposed method was validated against that using Doppler tissue imaging and some preliminary clinical experimental studies were presented to illustrate clinical values of the proposed method.

Colorizing grayscale images so that the resulting image appears natural is a hard problem. Previous colorization algorithms generally use just the luminance information and ignore the rich texture information, which means that regions with the same luminance but different textures may mistakenly be assigned the same color. A novel automatic texture-map-based grayscale image colorization method is proposed. The texture map is generated with bilateral decomposition and a Gaussian high pass filter, which is further optimized using statistical adaptive gamma correction method. The segmentation of the spatial map is performed using locally weighted linear regression on its histogram in order to match the grayscale image and the source image. Within each of the spatial segmentation, a weighted color-luminance correspondence is achieved by the results of locally weighted linear regression. The luminance-color correspondence between the grayscale image and the source image can thus be used to colorize the grayscale image directly. By considering the consistency of both color information and texture information between two images, various plausible colorization results are generated using this new method.

A joint image coding and image authentication scheme based on absolute moment block truncation coding (AMBTC) is proposed. In the proposed scheme, the authentication data is generated by using the pseudo random sequence. Then, the authentication codes are embedded into the bit maps of AMBTC-compressed image blocks. The embedded bit maps and these quantization levels are further losslessly compressed to cut down the required storage cost. Experimental results demonstrate that the proposed scheme achieves good image quality of the embedded image while keeping good detecting accuracy.

Satellite images are corrupted by noise during image acquisition and transmission. The removal of noise from the image by attenuating the high-frequency image components removes important details as well. In order to retain the useful information, improve the visual appearance, and accurately classify an image, an effective denoising technique is required. We discuss three important steps such as image denoising, resolution enhancement, and classification for improving accuracy in a noisy image. An effective denoising technique, hybrid directional lifting, is proposed to retain the important details of the images and improve visual appearance. The discrete wavelet transform based interpolation is developed for enhancing the resolution of the denoised image. The image is then classified using a support vector machine, which is superior to other neural network classifiers. The quantitative performance measures such as peak signal to noise ratio and classification accuracy show the significance of the proposed techniques.

An image watermarking scheme that combines Hermite functions expansion and space/spatial-frequency analysis is proposed. In the first step, the Hermite functions expansion is employed to select busy regions for watermark embedding. In the second step, the space/spatial-frequency representation and Hermite functions expansion are combined to design the imperceptible watermark, using the host local frequency content. The Hermite expansion has been done by using the fast Hermite projection method. Recursive realization of Hermite functions significantly speeds up the algorithms for regions selection and watermark design. The watermark detection is performed within the space/spatial-frequency domain. The detection performance is increased due to the high information redundancy in that domain in comparison with the space or frequency domains, respectively. The performance of the proposed procedure has been tested experimentally for different watermark strengths, i.e., for different values of the peak signal-to-noise ratio (PSNR). The proposed approach provides high detection performance even for high PSNR values. It offers a good compromise between detection performance (including the robustness to a wide variety of common attacks) and imperceptibility.

Object tracking is widely used in many applications such as intelligent surveillance, scene understanding, and behavior analysis. Graph-based semisupervised learning has been introduced to deal with specific tracking problems. However, existing algorithms following this idea solely focus on the pairwise relationship between samples and hence could decrease the classification accuracy for unlabeled samples. On the contrary, we regard tracking as a one-class classification issue and present a novel graph-based semisupervised tracker. The proposed tracker uses linear neighborhood propagation, which aims to exploit the local information around each data point. Moreover, the manifold structure embedded in the whole sample set is discovered to allow the tracker to better model the target appearance, which is crucial to resisting the appearance variations of the object. Experiments on some public-domain sequences show that the proposed tracker can exhibit reliable tracking performance in the presence of partial occlusions, complicated background, and appearance changes, etc.

A system is presented for spotting and searching keywords in handwritten Arabic documents. A slightly modified dynamic time warping algorithm is used to measure similarities between words. Two sets of features are generated from the outer contour of the words/word-parts. The first set is based on the angles between nodes on the contour and the second set is based on the shape context features taken from the outer contour. To recognize a given word, the segmentation-free approach is partially adopted, i.e., continuous word parts are used as the basic alphabet, instead of individual characters or complete words. Additional strokes, such as dots and detached short segments, are classified and used in a postprocessing step to determine the final comparison decision. The search for a keyword is performed by the search for its word parts given in the correct order. The performance of the presented system was very encouraging in terms of efficiency and match rates. To evaluate the presented system its performance is compared to three different systems. Unfortunately, there are no publicly available standard datasets with ground truth for testing Arabic key word searching systems. Therefore, a private set of images partially taken from Juma’a Al-Majid Center in Dubai for evaluation is used, while using a slightly modified version of the IFN/ENIT database for training.

We propose several methods to transplant the compound chaotic image encryption scheme with permutation based on three-dimensional (3-D) baker onto image formats such as the joint photographic experts group (JPEG) and graphics interchange format (GIF). The new methods avert the discrete cosine transform and quantization, which result in floating point precision loss, and succeed to encrypt and decrypt JPEG images lossless. The ciphered JPEG images generated by our solution own much better randomness than most other existing schemes. Our proposed method for GIF keeps the property of animation successfully. The security test results indicate the proposed methods have high security, and the speed of our algorithm is faster than classical solutions. Since JPEG and GIF image formats are popular contemporarily, we show that the prospect of chaotic image encryption is promising.

Although sparse representation (SR) based on the l¹-norm and l²-norm have achieved promising classification results for face recognition (FR) from frontal views, they both require an overcomplete training dictionary, which is usually unrealistic. We focus on addressing the problem of performing FR with SR with an incomplete dictionary. Motivated by the fact that image gradients could explicitly consider the relationships between neighboring pixel points and be less sensitive to illumination than image pixels, we introduce image gradients to SR and propose gradient-based sparse representation classification (GSRC). By combining image pixels and image gradients, GSRC has less model error and requires fewer training samples from each individual than sparse representation–based classification (SRC). Furthermore, GSRC can easily be combined with dimensionality reduction algorithms and be solved by the regularized least-square method, which makes GSRC work much faster than SRC. Extensive experimental results demonstrate that GSRC is quite efficient for both incomplete dictionary and occlusion and has a reasonable speed.

We deal with the problem of blind parameter estimation of signal-dependent noise from mono-component image data. Multispectral or color images can be processed in a component-wise manner. The main results obtained rest on the assumption that the image texture and noise parameters estimation problems are interdependent. A two-dimensional fractal Brownian motion (fBm) model is used for locally describing image texture. A polynomial model is assumed for the purpose of describing the signWe deal with the problem of blind parameter estimation of signal-dependent noise from mono-component image data. Multispectral or color images can be processed in a component-wise manner. The main results obtained rest on the assumption that the image texture and noise parameters estimation problems are interdependent. A two-dimensional fractal Brownian motion (fBm) model is used for locally describing image texture. A polynomial model is assumed for the purpose of describing the signal-dependent noise variance dependence on image intensity. Using the maximum likelihood approach, estimates of both fBm-model and noise parameters are obtained. It is demonstrated that Fisher information (FI) on noise parameters contained in an image is distributed nonuniformly over intensity coordinates (an image intensity range). It is also shown how to find the most informative intensities and the corresponding image areas for a given noisy image. The proposed estimator benefits from these detected areas to improve the estimation accuracy of signal-dependent noise parameters. Finally, the potential estimation accuracy (Cramér-Rao Lower Bound, or CRLB) of noise parameters is derived, providing confidence intervals of these estimates for a given image. In the experiment, the proposed and existing state-of-the-art noise variance estimators are compared for a large image database using CRLB-based statistical efficiency criteria. al-dependent noise variance dependence on image intensity. Using the maximum likelihood approach, estimates of both fBm-model and noise parameters are obtained. It is demonstrated that Fisher information (FI) on noise parameters contained in an image is distributed nonuniformly over intensity coordinates (an image intensity range). It is also shown how to find the most informative intensities and the corresponding image areas for a given noisy image. The proposed estimator benefits from these detected areas to improve the estimation accuracy of signal-dependent noise parameters. Finally, the potential estimation accuracy (Cramér-Rao Lower Bound, or CRLB) of noise parameters is derived, providing confidence intervals of these estimates for a given image. In the experiment, the proposed and existing state-of-the-art noise variance estimators are compared for a large image database using CRLB-based statistical efficiency criteria.

When the horizon or long edges are skewed in photos, they may seem unstable unless they are artistic intentions, and hence we may wish to correct the skews. For the skew correction of faint as well as strong horizons, we propose a skew estimation method for natural images. We first apply a long-block-based edge detector that can construct edge maps even when the edge is faint and/or background is cluttered. We also propose a robust line-detection method that uses the generated edge map, based on progressive probabilistic Hough transform followed by refinement steps. For each of the detected lines, we define their weight and estimate the image skew based on the weighted votes from the lines. Since all the pixels in the long-blocks are used for the edge-map construction, the proposed method can find noisy or faint lines while rejecting curved or short lines. Experimental results show that the first salient angle corresponds with the image skew in most cases, and the skews are successfully corrected.

Based on the curve evolution and the level set method, we proposed a geometric active contour model in a partial differential equation formulation. The evolution PDE is directly derived from the mean curvature motion with a transition region-based external force, and then is further simplified to a nonhomogeneous linear diffusion equation. Due to the desired sign property of the external force, the level set function starting with an arbitrary bounded function can adaptively move up or down in accordance with image information and has finally the opposite sign inside and outside the objects. The experiments show that the proposed model is very appropriate for a wider range of images, especially for real images with high noise and blurred boundaries.

An affine registration algorithm for multidimensional point sets under the framework of Lie group is proposed. This algorithm studies the affine registration between two data sets, and puts the expectation maximization-iterative closest point (EM-ICP) algorithm into the framework of Lie group, since all affine transformations form a Lie transformation group. The registration is carried out via minimizing an energy functional depending on elements of the affine transformation Lie group. The key point for applying the idea of Lie group is that, during the minimization via iteration, we must guarantee the next iteration step of the transformation is still an element in the same group, starting from an element in a Lie group. Our solution is utilizing the element of Lie algebra to represent that of Lie group near the identity via the exponential map, i.e., we use the first canonical coordinate representation of Lie group. Several comparative experiments between the proposed Lie-EM-ICP algorithm and the Lie-ICP algorithm are performed, showing that the proposed algorithm is more accurate and robust, especially in the presence of outliers. This algorithm can also be generalized to other registration problems in general, provided that desired transformations are within certain Lie group.

To improve the antinoise performance of the smallest univalue segment assimilating nucleus (SUSAN) edge detector, a nonlocal means-based SUSAN edge detector is proposed. The proposed method first determines the initial SUSAN edge response based on the image patch convolved with an adaptive kernel instead of the single pixel. Then it computes the final edge response using the weighted sum of the initial edge responses of the pixels with their structures similar to the considered pixel. Extensive simulations on natural and real images demonstrate that compared with state-of-the-art detectors, the proposed method performs much better in terms of robustness to noise and edge detection and it provides significantly higher values of Pratt’s figure of merit and performance measure.

Mammogram acquisition in digital format is one of the most relevant steps for image processing in computer-aided detection schemes for mammography. We investigate film digitizer systems using different technologies to determine their influence on the results of mammography image segmentation schemes. It also provides image scanning process regardless of the technology by the development of automatic software based on the digitizers’ characteristic curves. Comparative assessment of digitizer properties and features is performed as well as the software for managing the digitized image acquisition. The images were obtained from six different digitizers and evaluated by means of statistical analysis. Tests were conducted for comparing the responses from each equipment, regarding their respective curves, and they have presented significant variations relative to the original characteristic curve of high quality films used as reference—which largely influence the performance of processing schemes applied on sets of mammography images digitized by those systems. However, when our proposed scanning software was applied with intensity transformation procedure based on the characteristic curve “correction,” the images were comparable to the film optical density, which has improved the processing technique’s performance. The results have pointed out it is possible to achieve high sensitivity and performance of such schemes even with low-cost digitizer systems since their quality characteristics are well known and the procedure herein proposed is used within the mammogram scanning process.

The ever-growing penetration of communication networks, digital and Internet technologies in our everyday lives has the transmission of text data, as well as multimedia data such as images and videos, possible. Digital images have a vast usage in a number of applications, including medicine and providing security authentication, for example. This applicability becomes evident when images, such as walking or people’s facial features, are utilized in their identification. Considering the required security level and the properties of images, different algorithms may be used. After key generation using logistic chaos signals, a scrambling function is utilized for image agitation in both horizontal and vertical axes, and then a block-chaining mode of operation may be applied to encrypt the resultant image. The results demonstrate that using the proposed method drastically degrades the correlation between the image components and also the entropy is increased to an acceptable level. Therefore, the image will become greatly resistant to differential attacks. However, the increasing scrambling rounds and the decreasing number of bits of the blocks result in increasing the entropy and decreasing the correlation.

We present an effective solution for unsupervised texture segmentation by taking advantage of the latent Dirichlet allocation (LDA) model. LDA is a generative topic model that is capable of hierarchically organizing discrete data including texts and images. We propose a new texture model by connecting texture primitives to the topic of LDA. The model is able to extract the characteristic features of a texture primitive and group them into a topic based on their frequencies of co-occurrence. Here, the feature descriptor is the connection of Haar-like features of multiple sizes. The segments of an image are finally obtained by identifying the homogeneous regions in the corresponding topic assignment map. The evaluation results for synthetic texture mosaics, remote sensing images, and natural scene images are illustrated.

Recently many computational photography applications need to decompose an image into a piecewise smooth base layer, containing large-scale variations in intensity, and a residual detail layer capturing the smaller-scale details in the image. In these applications, the image decomposition method requires multiscale ability to avoid visual artifacts. In this paper, we propose a new model of image decomposition that has the properties of edge-preserving and multiscale ability. Inspired by techniques in computational geometry and morphological image analysis, we use the α -scale space of the input image to extract information about oscillations. We define detail as oscillations between upper and lower envelope of the input image. Building on the key observation that the spatial scale of oscillations is characterized by the α value, we develop an algorithm for decomposing images into multiple scales of superposed oscillations. Compared with traditional image decomposition methods, our method has three advantages as follows: (1) precisely controls scale parameter; (2) preserves edge while decomposing; and (3) decouples noise layer from noisy image. Finally, we compare our results with current existing edge-preserving image decomposition algorithms and demonstrate applications.

An imager that can measure the distance from each pixel to the point on the object that is in focus at the pixel is described. This is accomplished by short photo-conducting lightguides at each pixel. In the eye the rods and cones are the fiber-like lightguides. The device uses ambient light that is only coherent in spherical shell-shaped light packets of thickness of one coherence length. Modern semiconductor technology permits the construction of lightguides shorter than a coherence length of ambient light. Each of the frequency components of the broad band light arriving at a pixel has a phase proportional to the distance from an object point to its image pixel. Light frequency components in the packet arriving at a pixel through a convex lens add constructively only if the light comes from the object point in focus at this pixel. The light in packets from all other object points cancels. Thus the pixel receives light from one object point only. The lightguide has contacts along its length. The lightguide charge carriers are generated by the light patterns. These light patterns, and thus the photocurrent, shift in response to the phase of the input signal. Thus, the photocurrent is a function of the distance from the pixel to its object point. Applications include autonomous vehicle navigation and robotic vision. Another application is a crude teleportation system consisting of a camera and a three-dimensional printer at a remote location.

A concept of image quality metamerism as an expansion of conventional metamerism defined in color science is introduced, and it is applied to segment similar color areas in a cultural property. The image quality metamerism can unify different image quality attributes based on an index showing the degree of image quality metamerism proposed. As a basic research step, the index is consisted of color and texture information and examined to investigate a cultural property. The property investigated is a pair of folding screen paintings that depict the thriving city of Kyoto designated as a nationally important cultural property in Japan. Gold-colored areas painted by using high granularity colorants compared with other color areas are evaluated based on the image quality metamerism index locally, then the index is visualized as a map showing the possibility of the image quality metamer to the reference pixel set in the same image. This visualization means a segmentation of areas where color is similar but texture is different. The experimental result showed that the proposed method was effective to show areas of gold color areas in the property.

Face recognition is a rapidly growing research area, which is based heavily on the methods of machine learning, computer vision, and image processing. We propose a rotation and noise invariant near-infrared face-recognition system using an orthogonal invariant moment, namely, Zernike moments (ZMs) as a feature extractor in the near-infrared domain and spectral regression discriminant analysis (SRDA) as an efficient algorithm to decrease the computational complexity of the system, enhance the discrimination power of features, and solve the “small sample size” problem simultaneously. Experimental results based on the CASIA NIR database show the noise robustness and rotation invariance of the proposed approach. Further analysis shows that SRDA as a sophisticated technique, improves the accuracy and time complexity of the system compared with other data reduction methods such as linear discriminant analysis.

We present an evolutionary adaptive eye-tracking framework aiming for low-cost human computer interaction. The main focus is to guarantee eye-tracking performance without using high-cost devices and strongly controlled situations. The performance optimization of eye tracking is formulated into the dynamic control problem of deciding on an eye tracking algorithm structure and associated thresholds/parameters, where the dynamic control space is denoted by genotype and phenotype spaces. The evolutionary algorithm is responsible for exploring the genotype control space, and the reinforcement learning algorithm organizes the evolved genotype into a reactive phenotype. The evolutionary algorithm encodes an eye-tracking scheme as a genetic code based on image variation analysis. Then, the reinforcement learning algorithm defines internal states in a phenotype control space limited by the perceived genetic code and carries out interactive adaptations. The proposed method can achieve optimal performance by compromising the difficulty in the real-time performance of the evolutionary algorithm and the drawback of the huge search space of the reinforcement learning algorithm. Extensive experiments were carried out using webcam image sequences and yielded very encouraging results. The framework can be readily applied to other low-cost vision-based human computer interactions in solving their intrinsic brittleness in unstable operational environments.

This research studied the effect of cross-talk and three-dimensional (3-D) cues upon viewers’ visual fatigue and image quality evaluation in both mirror-type 3-D displays and barrier-type autostereoscopic displays. The experiment included three parts. First, the effect of system cross-talk, shadow, and linear perspective of 3-D pictures, in a mirror-type 3-D display, was discussed. Second, an acceptable threshold level of system cross-talk was sought. Third, the fitting system cross-talk level, in barrier-type autostereoscopic displays corresponding to that in mirror-type 3-D displays, was determined. The results showed that system cross-talk, shadow, and linear perspective, which were chosen from the various depth cues to study in this research, were significantly affecting image quality evaluation when watching 3-D displays. The acceptable system cross-talk level in mirror-type 3-D displays ranged from 18% to 23%. The content of a 3-D picture was critical to influence perceived image quality of autostereoscopic displays. In this experiment, viewers did not get visual fatigue and visual discomfort when watching 3-D pictures for 10 min. Moreover, experts ranked higher than novices in evaluating 3-D image quality. The outcomes of this study could be served as a reference for the design of 3-D displays. As for future work, an interesting direction to explore will be the effect of the dynamic 3-D films on viewing experience.

A new technique is presented for enhancing the contrast in digital images, combining the theory of partitioned iterated function system (PIFS) and image segmentation. The image is first segmented through the region growing segmentation technique, and the PIFS enhancement algorithm is applied separately to each image segment. The defined PIFS of each section is modeled by a contractive transformation, which consists of an affine spatial transform, as well as the linear transform of the graylevels of image segment pixels. The transformation of the graylevels is determined by two parameters that adjust the brightness and contrast of the transformed image segment. After the PIFS algorithm is applied to each extracted image segment, a lowpass version of the original image is created. The contrast-enhanced image is obtained by suitably combining the original image with its lowpass version. The proposed regional PIFS approach was applied to numerous test images, ranging from medical data of various modalities to standard images. The obtained quantitative and qualitative results showed superior performance on behalf of the proposed method when compared with three other widely used contrast enhancement methods, namely, contrast stretching, unsharp masking, and contrast-limited adaptive histogram equalization.

We present invariant image watermarking based on a recently introduced set of polar harmonic transforms and angular radial transforms and their comparative analysis with state-of-art approaches based on Zernike moments and pseudo-Zernike moments (ZMs/PZMs). Similar to ZMs/PZMs, these transforms provide rotation invariance and resilience to noise while mitigating inherent limitations like numerical instability and computational cost at high order of moments. These characteristics motivate us to design invariant transform-based invariant image watermarking schemes that can withstand various intentional or unintentional attacks, handle large bitcarriers, and work in a limited computing environment. A comparative performance evaluation of watermarking systems regarding critical parameters like visual imperceptibility, embedding capacity, and watermark robustness against geometric transformations, common signal processing distortions, and Stirmark attacks is performed along with the empirical analysis of various inherent properties of transforms and moments such as magnitude invariance, reconstruction capabilities, and computational complexity to investigate relationships between the performance of watermarking schemes and inherent properties of transforms.

An improved means of finding the direction of a light source illuminating a sphere is presented. In particular, the vertical slant angle of a light source is computed using a new approach. The planar tilt angle of a source is known to be related to the average of the local gradient vectors of an image of a lit sphere. This research uncovers a straightforward link between the length of the average of the local normalized gradient vectors of an image and the slant angle of the corresponding light source. The new technique is far more effective when compared with existing approaches over a variety of test images.

We propose an image encryption/decryption algorithm based on chaotic control parameter and hyperchaotic system with the composite permutation-diffusion structure. Compound chaos mapping is used to generate control parameters in the permutation stage. The high correlation between pixels is shuffled. In the diffusion stage, compound chaos mapping of different initial condition and control parameter generates the diffusion parameters, which are applied to hyperchaotic cellular neural networks. The diffusion key stream is obtained by this process and implements the pixels’ diffusion. Compared with the existing methods, both simulation and statistical analysis of our proposed algorithm show that the algorithm has a good performance against attacks and meets the corresponding security level.

In this study, a practical forensics aided steganalyzer (FA-steganalyzer) for heterogeneous bitmap images is constructed, which can properly handle steganalysis problems of mixed image sources including raw single-sampled images and resampled images at different scaling factors. The proposed FA-steganalyzer consists of a resampling detector followed by two corresponding steganalyzers, one of which is used to deal with raw single-sampled images and the other is used for mixed resampled images at different scaling factors. Extensive experimental results show that the proposed FA-steganalyzer outperforms the existing steganalyzer that is trained on a mixed dataset especially at low embedding rates.

An anisotropic optical flow estimation method based on self-adaptive cellular neural networks (CNN) is proposed. First, a novel optical flow energy function which contains a robust data term and an anisotropic smoothing term is projected. Next, the CNN model which has the self-adaptive feedback operator and threshold is presented according to the Euler–Lagrange partial differential equations of the proposed optical flow energy function. Finally, the elaborate evaluation experiments indicate the significant effects of the various proposed strategies for optical flow estimation, and the comparison results with the other methods show that the proposed algorithm has better performance in computing accuracy and efficiency.

This book serves as an introduction to the flourishing field of super-resolution imaging. It is a compiled volume, with different authors for each of its 14 chapters. While not having a strong outline or textbook format, the chapters group into several sections.