Adaptive Equalization Technique Research Articles

Developing Learning-Based Preprocessing Methods for Detecting Complicated Vehicle Licence Plates

A licence plate detection (LPD) system is an important tool in several roadway traffic applications. This study aims to develop an advanced detection system that works well in complicated scenarios. It proposes a robust preprocessing enhancement method for accurately detecting the licence plates from difficult vehicle images. The proposed method includes the combination of a Gaussian filter, an enhancement cumulative histogram equalization method, and a contrast-limited adaptive histogram equalization technique. The local binary pattern and median filter with histogram of oriented gradient descriptors are used as powerful tools to extract key features from three types of licence plate resolutions. The extracted features are used as input to support vector machine classifier. Processing methods, such as a position-based method are used with the detector to reduce unwanted bounding boxes, as well as false positive values. Four databases consisting of 2050 vehicle images under different conditions are used. Various detection metrics, object localization, and the receiver operating characteristic (ROC) curve are used to evaluate the performance of the proposed method. The experimental results on vehicles databases in several languages, including English, Chinese, and Arabic number plates, show that the proposed method has achieved significant performance improvements. It outperforms the state-of-the-art approaches in terms of both the detection rate and the processing time. The detection rate when trained with 1520 LP images is 99.62% with a false positive rate of 1.675% for complicated images. The average detection time per vehicle image is 0.2408 milliseconds.

Adaptive Nonlinear Equalizer for Full-Duplex Underwater Acoustic Systems

To enable full-duplex (FD) operation in underwater acoustic systems, the strong self-interference (SI) from the near-end transmitter should be efficiently cancelled. Digital cancellation is considered as the main approach for the SI cancellation. It is believed that the key challenge in achieving a high level of cancellation is to overcome the nonlinear distortion in the transmit and receive chains. The majority of the nonlinearity is introduced by the power amplifier (PA), which can be dealt with by using the PA output as the reference signal for digital cancellation. Further nonlinearity comes mainly from the hydrophone pre-amplifier. For applications working in half-duplex mode, the pre-amplifier can normally be assumed to be linear due to the low received signal level. For FD operations, the strong SI might result in a nonlinear response of the pre-amplifier, and this nonlinearity should be equalized to achieve a high cancellation performance. In this paper, we propose a technique for equalizing the nonlinearity in the pre-amplifier. This is done using a basis expansion model of the nonlinear equalizer response. More specifically, the Legendre polynomials are used as the basis functions. The expansion coefficients are updated to reduce the mean squared error or a cost function derived based on the power spectrum of the received signal. The performance of the equalizer is evaluated using the experimental data with artificial nonlinear distortion and with real nonlinearity introduced by the hydrophone pre-amplifier. Both sets of results demonstrate that nonlinear distortions can be effectively equalized using the proposed adaptive equalization technique.

Localization and Classification of Rice-grain Images Using Region Proposals-based Convolutional Neural Network

This paper proposes a solution to localization and classification of rice grains in an image. All existing related works rely on conventional based machine learning approaches. However, those techniques do not do well for the problem designed in this paper, due to the high similarities between different types of rice grains. The deep learning based solution is developed in the proposed solution. It contains pre-processing steps of data annotation using the watershed algorithm, auto-alignment using the major axis orientation, and image enhancement using the contrast-limited adaptive histogram equalization (CLAHE) technique. Then, the mask region-based convolutional neural networks (R-CNN) is trained to localize and classify rice grains in an input image. The performance is enhanced by using the transfer learning and the dropout regularization for overfitting prevention. The proposed method is validated using many scenarios of experiments, reported in the forms of mean average precision (mAP) and a confusion matrix. It achieves above 80% mAP for main scenarios in the experiments. It is also shown to perform outstanding, when compared to human experts.

Enhancement of MRI Prostate Image Using Histogram Equalization Techniques

In this cutting edge world, Medical image processing in computerized field needs a compelling MRI image modality with less commotion and improved contrast of image. This is conceivable by utilizing image enhancement methodology. Image enhancement is referenced as a system of changing or altering image so as to make it progressively sensible for explicit applications and is utilized to enhance or improve contrast proportion, splendor of image, expel clamor from image and make it less hard to perceive. The purpose behind inclining toward Medical Resonance Imaging (MRI) is that it is a mind boggling medical technology which gives more useful information regarding malignancy, stroke and various another ailments. It helps the doctors to distinguish the diseases more adequately. MRI has exceptionally low difference proportion. To improve the contrast of MRI image, we utilized Histogram equalization technique. In which, Histogram Equalization, Local Histogram Equalization, Adaptive Histogram Equalization and Contrast Limited Adaptive Histogram Equalization techniques were used and it is pondered.

Sparse Direct Adaptive Equalization for Single-Carrier MIMO Underwater Acoustic Communications

The sparse direct adaptive equalization technique recently received many attentions for single-carrier underwater acoustic communications. By taking advantage of the sparse (nonuniform) structure of a direct adaptive equalizer (DAE), one obtains a sparse DAE with improved performance and/or reduced complexity compared with its nonsparse counterpart. In this article, the sparse DAE is revisited with two contributions made: First, a comprehensive comparison is made for existing sparse DAEs designed with the proportionate-updating (PU) or the zero-attracting (ZA) adaptive filtering principles. The comparison concludes that the PU-type DAE outperforms the ZA-type DAE in terms of complexity, performance, and operability, thus shall be favored in practical use. Moreover, it motivates the design of a sparse DAE, named the selective ZA improved proportionate normalized least mean squares DAE (SZA-IPNLMS-DAE), based on the combination of the PU and ZA principles. The SZA-IPNLMS-DAE outperforms existing sparse DAEs armed with only one sparsity-promoting mechanism; second, a partial tap update (PTU) scheme via hard thresholding is introduced to sparse DAEs for reducing their complexity without sacrificing performance. The resulting low-complexity and high-performance sparse PTU-DAE schemes are strong candidates for single-carrier UWA communications. Experimental results of multiple-input multiple-output UWA communications are presented to corroborate all above claims.

Adaptive Equalization for Dispersion Mitigation in Multi-Channel Optical Communication Networks

Optical communication networks (OCNs) provide promising and cost-effective support for the ultra-fast broadband solutions, thus enabling them to address the ever growing demands of telecommunication industry such as high capacity and end users’ data rate. OCNs are used in both wired and wireless access networks as they offer many advantages over conventional copper wire transmission such as low power consumption, low cost, ultra-high bandwidth, and high transmission rates. Channel effects caused by light propagation through the fiber limits the performance, hence the data rate of the overall transmission. To achieve the maximum performance gain in terms of transmission rate through the OCN, an optical downlink system is investigated in this paper using feed forward equalizer (FFE) along with decision feedback equalizer (DFE). The simulation results show that the proposed technique plays a key role in dispersion mitigation in multi-channel optical transmission to uphold multi-Gb/s transmission. Moreover, bit error rate (BER) and quality factor (Q-factor) below 10 − 5 and above 5, respectively, are achieved with electrical domain equalizers for the OCN in the presence of multiple distortion effects showing the effectiveness of the proposed adaptive equalization techniques.

Mathematical Morphology and Optimum Principal Curvature Based Segmentation of Blood Vessels in Human Retinal Fundus Images

The retinal abnormalities and diagnosis of Diabetic Retinopathy (DR), Glaucoma are accomplished by extraction of vessel network in human retinal images. An accurate segmentation is required for the pathological analysis. Various researchers proposed many automated systems for vessel segmentation, still this process needs an improvement due to the presence of abnormalities, different magnitude, dimension of the vessels, non-uniform lighting and variable structure of the retina. The proposed work is a new method for retinal vessel segmentation, which consists of three phases, (i) The vessels network is enhanced by using Contrast Limited Adaptive Histogram Equalization(CLAHE) and Median filtering techniques (ii) the smoothened image is segmented based on mathematical morphology and maximum principal curvature followed by cleaning operation to remove the small objects, (iii) the segmented image is compared with hand labeled Ground Truth image and is evaluated with the True Positive, False Positive , True Negative and False Negative parameters. The performance of this work is tested with the images existing in DRIVE database. This work achieves 0.965 Accuracy, 0.752 Sensitivity and 0.989 Specificity.

Detection and Segmentation of Cancer Regions in Oral MRI images using ANFIS Classification Method

The detection of oral cancer is an important area of research in the literature today. About 82% of the patients are diagnosed at the early stage and 27% of the patients are diagnosed at the advanced stage. Early detection reduces the mortality rate of the patients. An automated approach is proposed to detect and segment the oral cancer in oral Magnetic Resonance Images (MRI). The quality of the image is improved using adaptive mean filter and enhanced using adaptive histogram equalization technique. The enhanced image is transformed using Gabor transform and the features of the oral image are extracted from this transformed image. These features are classified using ANFIS classification approach. Morphological approaches are used to segment the cancer region in the classified abnormal oral MRI images.

A PSO-Based Hybrid Adaptive Equalization Algorithm for Asynchronous Cooperative Communications

This paper proposes an adaptive equalization algorithm for asynchronous cooperative communications in ad hoc networks, where amplify-and-forward relays are adopted, each of which is equipped with single antenna. Adaptive equalization technique is carried out at the destination node in this paper to remove inter-symbol interference, which is caused by the retransmissions of the asynchronous relays. Least mean squares (LMS) has been regarded as an effective adaptive method, but it has difficulty in obtaining the optimal solution. In this paper, we present a hybrid adaptive scheme by combining particle swarm optimization (PSO) with conventional LMS algorithm, where PSO is utilized to search the optimal solution during the iterative process, and LMS is employed to avoid the local convergence, which is usually caused in PSO. Numerical simulation results show that, the proposed scheme outperforms conventional LMS algorithm in convergence performance over Rayleigh flat fading channel, and meanwhile, a signal–noise-ratio gain of 6 dB or so is obtained when BER is 10−3.

Feature extraction and segmentation of medical images for MRI and Digital mammogram

Developing a computer aided diagnosis system (CAD) is an extremelychallenging task. One of the major goals of CAD is to help the radiologist to makegood decisions by detecting and analyzing characteristics of benign and malignant lesions. In this context, we present accurate and automatic method that, detect and extract malignancy descriptors of breast and meningioma brain tumor.Our applied an algorithm that uses enhancement image based on homomorphicfiltering and adaptive histogram equalization technique. This work was proposedby Zhang Chaofu et al. [3]. A region of interest is determinated using K meansclustering. And then, we employed basically wavelet transform to extract pertinent features for meningioma tumor, geometric and texture characteristics forbreast tumor in order to classify malignancy lesion.

RETRACTED ARTICLE: An enhanced diabetic retinopathy detection and classification approach using deep convolutional neural network

The objective of this study is to propose an alternative, hybrid solution method for diagnosing diabetic retinopathy from retinal fundus images. In detail, the hybrid method is based on using both image processing and deep learning for improved results. In medical image processing, reliable diabetic retinopathy detection from digital fundus images is known as an open problem and needs alternative solutions to be developed. In this context, manual interpretation of retinal fundus images requires the magnitude of work, expertise, and over-processing time. So, doctors need support from imaging and computer vision systems and the next step is widely associated with use of intelligent diagnosis systems. The solution method proposed in this study includes employment of image processing with histogram equalization, and the contrast limited adaptive histogram equalization techniques. Next, the diagnosis is performed by the classification of a convolutional neural network. The method was validated using 400 retinal fundus images within the MESSIDOR database, and average values for different performance evaluation parameters were obtained as accuracy 97%, sensitivity (recall) 94%, specificity 98%, precision 94%, FScore 94%, and GMean 95%. In addition to those results, a general comparison of with some previously carried out studies has also shown that the introduced method is efficient and successful enough at diagnosing diabetic retinopathy from retinal fundus images. By employing the related image processing techniques and deep learning for diagnosing diabetic retinopathy, the proposed method and the research results are valuable contributions to the associated literature.

Single-sample face recognition under varying lighting conditions based on logarithmic total variation

The logarithmic total variation (LTV) algorithm is a classical algorithm that is proposed to address the illumination interference in face recognition. Some state-of-the-art techniques based on LTV assume that the illumination component mainly lies in the low-frequency features among face images. However, these techniques adopt unsuitable methods to process low-frequency features, resulting in final unsatisfactory recognition rates. In this paper, we propose an improved illumination normalization method based on the LTV method, called the RETINA&TH-LTV algorithm. In this algorithm, the retina model is utilized to eliminate most of the illumination component in low-frequency features. Then, an advanced contrast-limited adaptive histogram equalization technique is proposed to remove the residual lighting component. At the same time, through realizing threshold-value filtering on high-frequency features, the enhancement of facial features is achieved. Finally, the processed frequency features are combined to form a robust holistic feature image, which is then utilized for recognition. Insufficient training images in face recognition are also taken into consideration in this research. Comparative experiments for single-sample face recognition are conducted on YALE B, CMU PIE and our self-built driver databases. The nearest neighbor classifier and extended sparse representation classifier are employed as classification methods. The results indicate that the RETINA&TH-LTV algorithm has promising performance, especially in serious illumination and insufficient training sample conditions.

An Efficient Framework to Detect Cracks in Rail Tracks Using Neural Network Classifier

The detection of defects or cracks in rail track plays an important role in railway management, which prevents train accidents in both summer and rainy seasons. During summer, the cracks are formed on the track which slips the train wheel. In rainy environment, the rail tracks are affected by corrosion which also produced cracks on it. Methods: In present method, the cracks or defects are detected Echo image display device or semi conduction magnetism sensor devices which consumes more time. The proposed method enhances the track image using adaptive histogram equalization technique and further features as Grey Level Co-occurrence Matrix (GLCM) and Local Binary Pattern (LBP) feature are extracted from the enhanced rail track image. These extracted features are trained and classified using neural network classifier which classifies the rail track image into either cracked or non-cracked image. The novelty of this work is to use soft computing approach for the detection of cracks in rail tracks. This methodology is trained by several crack images which are obtained from different environment. This method automatically classifies the current image based on the trained patterns, thus improves the classification accuracy. Findings: The performance of the proposed system achieves the accuracy rate of 94.9%, with respect to manually crack detected and segmented images.

Subband Index Carrierless Amplitude and Phase Modulation for Optical Communications

Subband index carrierless amplitude and phase modulation (SI-CAP) is proposed and investigated in this paper to improve the performance of multi-band CAP ( $m$ -CAP). The bit-error-rate (BER) analysis of the proposed SI-CAP is derived and verified through computer simulations. A detection scheme is also developed for SI-CAP, which achieves maximum likelihood performance at a significantly lower complexity. In addition, an experimental demonstration of the proposed SI-CAP is carried out for optical communication systems. Furthermore, an adaptive equalization technique is implemented to further enhance the performance gain of SI-CAP. It is shown that for the same order of complexity, the SI-CAP exhibits higher spectral and energy efficiency in comparison to $m$ -CAP. Therefore, the performance gain of SI-CAP along with its design flexibility make it a suitable candidate for optical communication systems.

An approach for de-noising and contrast enhancement of retinal fundus image using CLAHE

Now-a-days medical fundus images are widely used in clinical diagnosis for the detection of retinal disorders. Fundus images are generally degraded by noise and suffer from low contrast issues. These issues make it difficult for ophthalmologist to detect and interpret diseases in fundus images. This paper presents a noise removal and contrast enhancement algorithm for fundus image. Integration of filters and contrast limited adaptive histogram equalization (CLAHE) technique is applied for solving the issues of de-noising and enhancement of color fundus image. The efficacy of the proposed method is evaluated through different performance parameters like Peak Signal to Noise Ratio (PSNR), Structural Similarity Index (SSIM), Correlation coefficient (CoC) and Edge preservation index (EPI). The proposed method achieved 7.85% improvement in PSNR, 1.19% improvement in SSIM, 0.12% improvement in CoC and 1.28% improvement in EPI when compared to the state of the art method.

A Hybrid Particle Swarm Optimization Technique for Adaptive Equalization

Adaptive equalization mitigates the distortions caused by radio channels. The least mean square (LMS) and the recursive least squares (RLS) algorithms are used for such purpose. Recently, particle swarm optimization (PSO) algorithms such as PSO using a linear time decreasing inertia weight (PSO-W) and the PSO using constant constriction factor (PSO-CCF) were shown to be very effective in handling systems having nonlinear behavior. However, these algorithms can be trapped in local minima. This paper presents a new PSO-based algorithm called the hybrid PSO (HPSO) that is capable to handle such problems. The HPSO includes the randomization of particles to improve the search capacity of the swarm, which in turn reduces the probability of being trapped in some local minima. It also adapts the inertia weight assignment to the particles. Extensive simulation results are conducted to confirm the consistency in the performance of the HPSO algorithm in different scenarios. The proposed HPSO secures the minimum steady-state error as compared to LMS and other PSO-based algorithms in both nonlinear and linear channels. Finally, the proposed HPSO algorithm shows a great improvements in Bit Error Rate and convergence rate.

Kalman-MLSE Equalization for NLIN Mitigation

We investigate the potential of adaptive equalization techniques to mitigate interchannel nonlinear interference noise (NLIN). We derive a lower bound on the mutual information of a system using adaptive equalization, showing that the channel estimation error determines the equalizer's performance. We develop an adaptive equalization scheme which uses the statistics of the NLIN to obtain improved detection based on Kalman filtering and maximum likelihood sequence estimation. This scheme outperforms commonly used equalizers and significantly increases performance.

A Robust Technique for Edge Detection Using Integration of Entropy Threshold—Canny (InTEC)

Edge detection is one of the important stages in digital image processing and computer vision. In general, edge detection is a process of identifying and validating sudden discontinuities in pixel intensities for grayscale images. This paper focuses on discussing a robust technique in edge detection using integration of several well-known techniques. At first, contrast enhancement approach using contrast limited adaptive histogram equalization (CLAHE) technique was performed to increase the brightness of the images. An adaptive filtering process will be applied in order to reduce noisy elements. To achieve our aims on producing a robust technique, we synthesized Canny technique with modified entropy based method. Based on the conducted performance test, InTEC technique able to identify the edge easily even though the noise percentage increases up to 30%. The high recorded peak signal-to-noise ratio (PSNR) values showed that InTEC technique significantly robust as compared to other prominent techniques.

Digital Imaging Tool to Enhance Otolith Microstructure for Estimating Age in Days in Juvenile and Adult Fish

Age estimation based on fish otolith microstructure analysis is a very repetitive and time-consuming task. The lack of appropriate image analysis software, capable of both overlaying a number of images automatically and recording a high number of daily increments, has been a significant limitation in counting and measuring daily growth increments in large otoliths from juvenile and adult fish individuals. This paper presents a new software to assist marine biologists with faster, more efficient, and more reliable microstructure readings of fish otoliths. Open source code is preferred so that software packages can be updated with new image processing algorithms developed by the scientific community. The approach consists of three steps: 1) a single grayscale digital image combining images of different parts of the same fish otolith is obtained using the blind image registration technique fast normalized cross correlation; 2) the growth rings of the image are enhanced for age estimation purposes, using the adaptive histogram equalization technique; and 3) a semiautomatic interactive tool draws a simple polygonal chain along which the microstructures are easily identifiable and the points of interest can be marked, whose data will be saved automatically. This new tool opens up the opportunity of aging juvenile and adult fish individuals at regular intervals by counting growth rings in the otolith microstructure and facilitates working with other calcified pieces of marine species that exhibit a daily ring pattern, such as cephalopod beaks and mollusk shells.

Comparison of NRZ and duo-binary format in adaptive equalization assisted 10G-optics based 25G-EPON

We investigate and compare the requirements of FFE/DFE based adaptive equalization techniques for NRZ and Duo-binary based 25-Gb/s transmission, which are two of the most promising schemes for 25G-EPON. A 25-Gb/s transmission system based on 10G optical transceivers is demonstrated and the performance of only FFE and combination of FFE and DFE with different number of taps are compared with two modulation formats. The FFE/DFE based Duo-binary receiver shows better performance than NRZ receiver. For Duo-binary receiver, only 13-tap FFE is needed for BtB case and the combination of 17-tap FFE and 5-tap DFE can achieve a sensitivity of −23.45 dBm in 25 km transmission case, which is ∼0.6 dB better than the best performance of NRZ equalization. In addition, the requirements of training sequence length for FFE/DFE based adaptive equalization is verified. Experimental results show that 400 symbols training length is optimal for the two modulations, which shows a small packet preamble in up-stream burst-mode transmission.