Ezzatollah Salari

Successive elimination algorithm for motion estimation

The correspondence presents a fast exhaustive search algorithm for motion estimation. The basic idea is to obtain the best estimate of the motion vectors by successively eliminating the search positions in the search window and thus decreasing the number of matching evaluations that require very intensive computations. Simulation results demonstrate that although the performance of the proposed algorithm is the same as that using the exhaustive search, the computation time has been reduced significantly.

Texture segmentation using hierarchical wavelet decomposition

This paper presents a texture segmentation algorithm based on a hierarchical wavelet decomposition. Using Daubechies four-tap filter, an original image is decomposed into three detail images and one approximate image. The decomposition can be recursively applied to the approximate image to generate a lower resolution of the pyramid. The segmentation starts at the lowest resolution using the K-means clustering scheme and textural features obtained from various sub-bands. The result of segmentation is propagated through the pyramid to a higher resolution with continuously improving the segmentation. The lower resolution levels help to build the contour of the segmented texture, while higher levels refine the process, and correct possible errors.

Beamlet Transform‐Based Technique for Pavement Crack Detection and Classification

: This article presents a Beamlet transform-based approach to automatically detect and classify pavement cracks in digital images. The proposed method uses a pavement distress image enhancement algorithm to correct the nonuniform background illumination by calculating the multiplicative factors that eliminate the background lighting variation. To extract linear features such as surface cracks from the pavement images, the image is partitioned into small windows and a Beamlet transform-based algorithm is applied. The crack segments are then linked together and classified into four types: vertical, horizontal, transversal, and block. Simulation results show that the method is effective and robust in the extraction of cracks on a variety of pavement images.

A fast vector quantization encoding method for image compression

This paper presents a general search method to speed up the encoding process for vector quantization. The method exploits the topological structure of the codebook to dynamically eliminate the code vectors for encoding a particular input vector and thus decrease the number of distance calculations which require very intensive computations. The relations between the proposed method and several existing fast algorithms are discussed. Based on the proposed method, a new fast encoding algorithm for vector quantization is developed. Simulation results demonstrate that with little preprocessing and memory cost, the encoding time of the new algorithm has been reduced significantly while encoding quality remains the same with respect to exhaustive search. >

Pavement pothole detection and severity measurement using laser imaging

Over the years, Automated Image Analysis Systems (AIAS) have been developed for pavement surface analysis and management. The cameras used by most of the AIAS are based on Charge-Coupled Device (CCD) image sensors where a visible ray is projected. However, the quality of the images captured by the CCD cameras was limited by the inconsistent illumination and shadows caused by sunlight. To enhance the CCD image quality, a high-power artificial lighting system has been used, which requires a complicated lighting system and a significant power source. In this paper, we will introduce an efficient and more economical approach for pavement distress inspection by using laser imaging. After the pavement images are captured, regions corresponding to potholes are represented by a matrix of square tiles and the estimated shape of the pothole is determined. The vertical, horizontal distress measures, the total number of distress tiles and the depth index information are calculated providing input to a three-layer feed-forward neural network for pothole severity and crack type classification. The proposed analysis algorithm is capable of enhancing the pavement image, extracting the pothole from background and analyzing its severity. To validate the system, actual pavement pictures were taken from pavements both in highway and local roads. The experimental results demonstrated that the proposed model works well for pothole and crack detection.

Automated pavement distress detection using advanced image processing techniques

In this paper, a novel, fast and self-adaptive image processing method is proposed for the extraction and connection of break points of cracks in pavement images. The algorithm first finds the initial point of a crack and then determines the cracks classification into transverse, longitudinal and alligator types. Different search algorithms are used for different types of cracks. Then the algorithm traces along the crack pixels to find the break point and then connect the identified crack point to the nearest break point in the particular search area. The nearest point then becomes the new initial point and the algorithm continues the process until reaching the end of the crack. The experimental results show that this connection algorithm is very effective in maximizing the accuracy of crack identification.

Beamlet transform based technique for pavement image processing and classification

The goal of this paper is to automatically detect and classify pavement cracks using digital image processing techniques. This paper uses a pavement distress image enhancement algorithm to correct non-uniform background illumination by calculating the multiplication factors that eliminate the background lighting variations. To extract the linear features such as surface cracks from the pavement images a beamlet transform based algorithm is developed.

Image denoising using a neural network based non-linear filter in wavelet domain

Images are often corrupted as a result of various factors that can occur during acquisition and transmission processes. Image denoising is aimed at removing or reducing noise, so that a good-quality image can be obtained for various applications. The paper presents a neural network based denoising method implemented in the wavelet transform domain. A noisy image is first wavelet transformed into four subbands, then a trained layered neural network is applied to each subband to generate noise-removed wavelet coefficients from their noisy ones. The denoised image is thereafter obtained through the inverse transform on the noise-removed wavelet coefficients. Simulation results demonstrate that this method is very efficient in removing noise. Compared with other methods performed in the wavelet domain, it requires no a priori knowledge about the noise and needs only one level of signal decomposition to obtain very good denoising results.

Recognition of partially occluded objects using B-spline representation

This paper reports a method to recognize partially occluded objects using the B-spline representation of the boundary. Curve sgements are represented using B-splines which are piecewise polynomial curves guided by a sequence of points. The B-spline control points found from the boundary points is then used to extract local features of the curve. A Hough transform like method is applied to normalize the two curve boundaries using extracted local features. The merit of a match is evaluated using the normalized B-spline control points. The ability of the technique to handle partial boundary information is also demonstrated.

Optical material stress measurement using two orthogonally polarized sinusoidally intensity-modulated semiconductor lasers

A novel, compact, robust and highly versatile polarization-modulated electro-optical instrument for measuring material properties, fluid flow parameters, stress, strain and molecular structure of optically anisotropic materials has been developed in this paper. The new instrumentation uses two polarized laser beams. Each beam is linearly polarized with the two polarization states orthogonal to each other. The laser beams are sinusoidally intensity modulated with 180? phase difference by two laser drivers and a signal inverter connected to the output of one of the laser driver circuits. The anti-phase intensity modulation of each orthogonal polarization increases the instruments sensitivity through the use of heterodyning signal analysis techniques with a single lock-in amplifier?(LIA). When the two semiconductor laser beams are optically combined, the result produces a laser beam with a constant optical power level comprised of time-varying power levels in each orthogonal polarization state. The polarization state of the laser light is modulated without the use of a traditional modulator. The instrument photodetector produces a direct-current signal along with a periodic signal at the modulation frequency that is recovered by a LIA tuned to the modulation frequency. By combining these signals in the appropriate relationship, a materials phase retardance or average molecular orientation angle may be measured. The main advantages of this technique over existing methods are lower cost due to the lack of an optical modulator, small size when compared to a photoelastically modulated system and improved sensitivity over continuous-wave laser crossed-polarizer instruments.