Road crack is one of the prominent problems that can frequently occur in highways and main roads. The manual road crack evaluation is laborious, time-consuming, inaccurate, and it has several implementation issues. Conversely, the computer vision-based solution is very challenging due to the complex ambient conditions, including illumination, shadow, dust, and crack shape. Most of the cracks exist as irregular edge patterns and are the most important features for detection purpose. Recent advances in deep learning adopt a convolutional neural network as the base model to detect and localize crack with a single RGB image. Yet, this approach has an inaccurate boundary for crack localization, resulting in thicker and blurry edges. To overcome this problem, the study proposes a novel and robust road crack detection based on deep learning which also considers the original edge of the image as the additional feature. The main contribution of this work is adapting the original image gradient with the coarse crack detection result and refining it to produce more precise crack boundaries. Extensive experimental results have shown that the proposed method outperforms the former state-of-the-art methods in terms of the detection accuracy.

Normally, modeling 3D face is an effective way for pose-invariant recognition, yet its expensive computation significantly discourages potential applications. In this paper, a simple and fully automatic panoramic image-based pose-invariant face recognition method is proposed to present excellent accuracy with low complexity. In this paper, a face shape model with local morphing treatment is first constructed and considered as the alignment standard to deal with all of the possible geometric distortion problems. During the recognition phase, a proposed systematically designed algorithm with morphing and the selection function are both utilized to significantly ease the negative effects of various poses within ±45° in yaw and ±22.5° in pitch. As demonstrated in experimental results, a similar accuracy as that of the 3D start-of-the-arts is achieved with much less computational complexity.

Ocular recognition is expected to provide a higher flexibility in handling practical applications as oppose to the iris recognition, which only works for the ideal open-eye case. However, the accuracy of the recent efforts is still far from satisfactory at uncontrollable conditions, such as eye blinking which implies any poses of eyes. To address these issues, the skin texture, eyelids, and additional geometrical features are employed. In addition, to achieve higher accuracy, sequential forward floating selection is utilized to select the best feature combinations. Finally, the non-linear support vector machine is applied for identification purpose. Experimental results demonstrate that the proposed algorithm achieves the best accuracy for both open eye and blinking eye scenarios. As a result, it offers greater flexibility for the prospective subjects during recognition as well as higher reliability for security.

Processing efficiency can be a key factor which dominates the value of a commercial product and its value-added application such as security. In this study, a computationally reduced halftoning-based watermarking is proposed. In encoder, the Efficient Direct Binary Search (EDBS) is employed to generate reference table to ensure the output is in halftone format. Subsequently, a number of optimized compressive tables with various texture angles are established for subsequent table lookup. In decoder, the Least-Mean-Square (LMS) enlarges the differences among those phenotypes of the embedded angles and the required number of dimensions for each angle. Finally, the naïve Bayes classifier is employed to collect the possibility information for classifying various angles. As documented in the experimental results, good image quality and correct detect rate can be obtained simultaneously. Moreover, a high processing efficiency of 0.6 milliseconds for an image of size 512×512 can also be achieved, which can substantially increase the commercial competitive strength in printing market, in particular the security issue is well addressed.

In this paper, we present a projection-based medical augmented reality (AR) system for craniofacial application. With the aid of image registration and geometrical correction, the deformation of the projected image on a curved surface like cheek or forehead can be successfully recovered. From the experimental result, it is shown that the overlapped target region is up to 98% after the correction process and this demonstrates the effectiveness of the proposed system.

Error diffusion is an efficient halftone method for mainly being applied on printers. The promising high image quality and processing efficiency endorse it as a popular and competitive candidate in halftoning and multitoning applications. The multitoning is an extension of halftoning, adopting more than two-tone levels for the improvement of the similarity between an original image and the converted image. Yet, the banding effect, indicating the areas with discontinuous tone level, disturbs the visual perception, and thus seriously degrades image quality. To solve the banding effect, the tone-replacement strategy is proposed in this paper. As documented in the experimental results, excellent tone-similarity as that of the original image and promising reconstructed dot-distribution can be provided simultaneously. Comparing with the former banding-free methods, the apparent improvements/features suggest that the proposed method can be a very competitive candidate for multitoning applications.

The problem of front vehicle detection and tracking during the night, unlike during the day, is that the environment provides little visible information with which to use. However, because it is important task which can help drivers at night, we propose a novel algorithm to detect and track by the location of tail lights. Our algorithm calculates the intensity of the diffused tail lights, then uses this information to find and track the location of the vehicle. Experiments were performed using videos taken at night, and the results show that our algorithm has excellent performance.

A prior work proposed by Chung-Wu considered an edge-based lookup table to obtain good inversed image quality, yet it suffers from some drawbacks in terms of image quality, memory consumption, and complexity. In this correspondence, an improved scheme is proposed to deal with these issues.

In this study, a local contrast enhancement method, namely Parametric-Oriented Histogram Equalization (POHE), is proposed to effectively yield enhanced results. In general, the grayscale distribution of a specific region in an image can be modeled with a kernel function such as the Gaussian, and thus the corresponding estimated cumulative distribution function (cdf) can be considered as the transformation function for contrast enhancement. The required parameters, however, still need to access all of the pixels in the corresponding region, and thus consume a huge amount of computations. To cope with this, the concept of integral image is adopted to effectively derive the required parameters. In the experimental results, former well-known speed-oriented methods are adopted for comparison, and the results demonstrate that the proposed methods can provide high practical value for biometric and tracking/detection these active issues who desire high efficiency.

Look-up table (LUT) halftoning is an efficient way to construct halftone images and approximately simulate the dot distribution of the learned halftone image set. In this paper, a general mechanism named multiple look-up table (MLUT) halftoning is proposed to generate the halftones of direct binary search (DBS), whereas the high efficient characteristic of the LUT is still preserved. In the MLUT, the standard deviation is adopted as an important feature to classify various tables. In addition, the proposed quick standard deviation evaluation is employed to yield an extremely low computational complexity in calculating the standard deviation. In the parameter optimization, the autocorrelation is adopted because it can fully characterize the periodicity of dot distribution. Experimental results demonstrate that the dot distribution generated by the proposed method approximates to that of the DBS, which enables the proposed scheme as a very competitive candidate in the copying and printing industry.

In this paper, a classified-based post-compensation algorithm for Color Filter Array (CFA) demosaicing is proposed. This technique can be used for improving the image quality of the interpolated results obtained by other CFA images. First, each pixel is classified according to its neighborhood texture variance and angle. Then, different Least-Mean-Square (LMS) filters are trained to adopt for dealing pixels of various characteristics. As documented in the experimental results, the proposed scheme can substantially boost the image quality; in addition, a better visual perceptual can be obtained. Notably, the proposed method can be considered as effective post-compensation by applying for any former schemes to yield an even better image quality.

In this paper, a halftoning-based watermarking method is presented. This method enables high pixel-depth watermark embedding, while maintaining high image quality. This technique is capable of embedding watermarks with pixel depths up to 3 bits without causing prominent degradation to the image quality. To achieve high image quality, the parallel oriented high-efficient direct binary search (DBS) halftoning is selected to be integrated with the proposed orientation modulation (OM) method. The OM method utilizes different halftone texture orientations to carry different watermark data. In the decoder, the least-mean-square-trained filters are applied for feature extraction from watermarked images in the frequency domain, and the naïve Bayes classifier is used to analyze the extracted features and ultimately to decode the watermark data. Experimental results show that the DBS-based OM encoding method maintains a high degree of image quality and realizes the processing efficiency and robustness to be adapted in printing applications.

This study proposes a method which can generate high quality inverse halftone images from halftone images. This method can be employed prior to any signal processing over a halftone image or the inverse halftoning used in JBIG2. The proposed method utilizes the least-mean-square (LMS) algorithm to establish a relationship between the current processing position and its corresponding neighboring positions in each type of halftone image, including direct binary search, error diffusion, dot diffusion, and ordered dithering. After which, a referenced region called a support region (SR) is used to extract features. The SR can be obtained by relabeling the LMS-trained filters with the order of importance. Moreover, the probability of black pixel occurrence is considered as a feature in this work. According to this feature, the probabilities of all possible grayscale values at the current processing position can be obtained by the Bayesian theorem. Consequently, the final output at this position is the grayscale value with the highest probability. Experimental results show that the proposed method offers better visual quality than that of Mese-Vaidyanathan's and Chang et al's methods in terms of human-visual peak signal-to-noise ratio (HPSNR). In addition, the memory consumption is also superior to Mese-Vaidyanathan's method.

Former research on inverse halftoning most focus on developing a general-purpose method for all types of halftone patterns, such as error diffusion, ordered dithering, etc., while fail to consider the natural discrepancies among various halftoning methods. To achieve optimal image quality for each halftoning method, the classification of halftone images is highly demanded. This study employed the least mean-square filter for improving the robustness of the extracted features, and employed the naive Bayes classifier to verify all the extracted features for classification. Nine of the most well-known halftoning methods were involved for testing. The experimental results demonstrated that the classification performance can achieve a 100% accuracy rate, and the number of distinguishable halftoning methods is more than that of a former method established by Chang and Yu.

This work presents a novel reversible data-hiding scheme that embeds secret data into a transformed image and achieves lossless reconstruction of vector quantization (VQ) indices. The VQ compressed image is modified by the side-matched VQ scheme to yield a transformed image. Distribution of the transformed image is employed to achieve high embedding capacity and a low bit rate. Moreover, three configurations, under-hiding, normal-hiding, and over-hiding schemes, are utilized to improve the proposed scheme further for various applications. Experimental results demonstrate that the proposed scheme significantly enhances the compression ratio and embedding capacity. Experimental results also show that the proposed scheme achieves the best performance among approaches in literature in terms of the compression ratio and embedding capacity.

An algorithm for finding coupling points plays an important role in the Iterative Closest Point algorithm(ICP) which is widely used in registration applications in medical and 3-D architecture areas. In recent researches of finding coupling points, Approximate K-D tree search algorithm (AK-D tree) is an efficient nearest neighbor search algorithm with comparable results. We proposed Adaptive Dual AK-D tree search algorithm (ADAK-D tree) for searching and synthesizing coupling points as significant control points to improve the registration accuracy in ICP registration applications. ADAK-D tree utilizes AK-D tree twice in different geometrical projection orders to reserve true nearest neighbor points used in later ICP stages. An adaptive threshold in ADAK-D tree is used to reserve sufficient coupling points for a smaller alignment error. Experimental results are shown that the registration accuracy of using ADAK-D tree is improved more than the result of using AK-D tree and the computation time is acceptable.

This paper describes a novel drive status monitoring system. Two fixed cameras are used in the proposed system to capture the driver's face image and the front road image to find the driver's sight orientation, the lane path orientation, and the car heading direction, and these 3 orientation information are calculated for 2 correlation coefficients by using the soft-shape-context concept to indicate the driver's steering control degrees. Three driving situations of the safe zone, risky zone, and unconscious zone are suggested according the combined coefficient of 2 correlation coefficients. This proposed system is capable to cooperate with other pre-crash systems based on the distance information to neighbor cars.

A hybrid learning neuro-fuzzy system with asymmetric fuzzy sets (HLNFS-A) is proposed in this paper. The learning methods of random optimization (RO) and least square estimation (LSE) are used in hybrid way to train the system parameters of HLNFS-A to achieve stable and fast convergence. In the HLNFS-A, the premise and the consequent parameters are updated by RO and LSE, respectively. With the proposed asymmetric fuzzy sets (AFS), the neuro-fuzzy system can capture the essence of nonlinear property of dynamic system, when used in the application of modeling. To demonstrate the feasibility and the potential of the proposed approach, an example of chaotic time series for system identification and prediction is given to verify the nonlinear mapping capability of the HLNFS-A. The experimental results show that the proposed HLNFS-A can achieve excellent performance for system modeling.

A new compression scheme called the hybrid compression model (HCM) is proposed for compressing sets of similar images. The HCM employs the region growing technique to partition the median image of a set of similar images; and furthermore, it uses the centroid method to characterize the original image data. The differences between the predicted and the original image data are stored and encoded for later use. The efficacy of its application on progressive transmission of similar images over the networks is also studied. The experimental results on various images show that our method provides significant improvement in compression efficiency, ranging from 5.6% to 134.9% in comparison with traditional centroid methods.

This paper discusses thinning on 3D binary images with the 4-subfield approach. Although a thinning algorithm concerns binary images, the algorithm itself can be represented as a set of three-color reduction templates. A thinning algorithm is topology preserving if the set of all three-color templates is topology preserving. Sufficient and necessary conditions of time complexity O(n) were proposed for verifying the topological soundness of a 3D 4-subfield thinning algorithm of n three-color templates. Theories and techniques for computerizing such conditions were discussed. Two 4-subfield thinning algorithms on 3D images, one for generating medial curves, and the other one for generating medial surfaces, are proposed and proved to preserve topology by our sufficient and necessary conditions.

In this paper, a new feasible approach is proposed to determine the 3-D location and orientation as well as the radius of a cylinder pasted with a known-height label. The proposed approach uses a rectangle-shaped standard mark composed of two point-pairs for performing monocular image analysis. A monocular image of a labelled cylinder is first taken. Image processing and numerical analysis techniques are then used to set two point-pairs located on the boundary of the visible part of the cylindrical label. According to 3-D imaging geometry, the location and orientation as well as the radius of the cylinder relative to the camera are uniquely determined by using 3-D vector analysis and simple algebraic computation. Owing to the full linearity of the,derivation, this approach can achieve high speed requirement. In addition, no prerequisite restriction is imposed on the image-grabbing process and there is no need to perform the calibration of the focal length since it can be derived. Experimental results, with mean errors less than 3%, prove the feasibility of the proposed approach.

A feasible approach is proposed to determine the 3D location and orientation of a cylinder with a label stuck on it at a known height. The proposed approach uses a rectangle-shaped standard mark composed of two point-pairs for performing monocular image analysis. A monocular image of a labelled cylinder is first taken. Image processing and numerical analysis techniques are then used to select two point-pairs located on the boundary of the visible part of the cylindrical label. According to 3D imaging geometry, the location and orientation of the cylinder relative to the camera are uniquely determined by using 3D vector analysis and simple algebraic computation. Owing to the full linearity of the derivation, this approach can perform at high speed.>

A novel approach to robot guidance in an unfamiliar environment is presented. In previous guiding methods, a preinstalled map or predefined path is required for a robot navigating in its working space. The present method uses an indication post (IP) to provide the information on the destination, the direction, and the distance from the source point to the destination point, assuming that the mobile robot has no information concerning its workplace. To realize the concept, many IPs depicting simple information were designed. The content of IPs can be described by using a finite-state grammar. Image-processing, statistical, and syntactic pattern recognition approaches are integrated to solve the problems of IP finding, robot location determination, IP identification, and understanding. Experiments simulating practical environments were performed. The results verify that the IP candidates can be automatically found and the depicted information can be extracted and understood correctly, so that the mobile robot is able to plan a global optimal path from an arbitrary start point to an arbitrary end point.>