Equalization Filter Research Articles

Cotton-wool spots, red-lesions and hard-exudates distinction using CNN enhancement and transfer learning

<div>The automatic retinal disease diagnosis by artificial intelligent is an interesting and challenging topic in the medical field. It requires an appropriate image enhancement technique and a sufficient training dataset for the specific retina conditions. The aim of this study was to design an automatic diagnosis convolutional neural network (CNN) model which does not require a large training dataset to specifically identify diabetic retinopathy symptoms, which are cotton wool, exudates spots and red lesionin colour fundus pictures. A novel framework comprised image enhancement method by using upgraded contrast limited adaptive histogram equalization (UCLAHE) filter and transferred pre-trained networks was developed to classify the retinal diseases regarding to the symptoms. The performance of the proposed framework was evaluated based on accuracy, sensitivity and specificity metrics. The collected results have proven the robustness of the proposed framework in offering good accuracy in retina diseases diagnosis. </div>

INTELLIGENT sYSTEM FOR HUMAN AUTHENTICATION USING FUSION OF DORSAL HAND, PALM AND FINGER VEINS

Multimodal biometric systems roughly used to achieve extreme recognition accuracy. This paper reports a novel multimodal biometric system employing intelligent technique to authenticate human by fusion of dorsal hand, palm and finger veins pattern. We improved an image analysis technique to separate region of interest (ROI) from dorsal hand, palm and finger veins image. After separating ROI we construct a sequence of preprocessing steps to enhance dorsal hand, palm and finger veins images using Median filter, Wiener filter, Contrast Limited Adaptive Histogram Equalization (CLAHE) and Homomorphic filter to improve vein image. Our intelligent technique is based on the following intelligent algorithms, namely; principal component analysis (PCA) algorithm for feature extraction and k-Nearest Neighbors (K-NN) classifier for matching operation. The database selected was Bosphorus Hand Vein Database, CASIA Multi-Spectral Palmprint Image Database V1.0 (CASIA database) and the Shandong University Machine Learning and Applications - Homologous Multi-modal Traits (SDUMLA-HMT). The accomplished result for the fusion of three biometric traits was Correct Recognition Rate (CRR) is 99.21% with False Reject Rate (FRR) 0.04%.

Electric-to-acoustic pickup processing for string instruments: An experimental study of the guitar with a hexaphonic pickup.

A signal processing method to impart the response of an acoustic string instrument to an electric instrument that includes frequency-dependent string decay alterations is proposed. This type of processing is relevant when trying to make a less resonant instrument, such as an electric guitar, sound similar to a more resonant instrument, such as acoustic guitar. Unlike previous methods which typically only perform equalization, our method includes detailed physics-based string damping changes by using a time-varying filter which adds frequency-dependent exponential damping. Efficient digital filters are fit to bridge admittance measurements of an acoustic instrument and used to create equalization filters as well as damping correction filters. The damping correction filters are designed to work in real-time as they are triggered by onset and pitch detection of the signal measured through an under-saddle pickup to determine the intensity of the damping. A test case is presented in which an electric guitar is processed to model a measured acoustic guitar.

Chromatic dispersion equalization FIR filter design based on discrete least-squares approximation.

Chromatic dispersion (CD) equalization is one of the core tasks of the digital signal processing (DSP) chain in modern optical coherent receivers. A conventional impulse-invariant method for designing the CD equalization filter is revisited, improved by proper weighting, and reinterpreted as a Fourier series. To improve upon a direct evaluation of the passband least-squares (LS) approximation, we propose to design a CD equalization finite impulse response (FIR) filter based on a discrete LS approximation. The proposed method avoids numerical evaluation of nontrivial functions and relies only on Fourier transform. Its flexibility is corroborated by a filter design demonstration of joint matched filtering and CD equalization.

Intelligent Equalization Based on RBF LSSVM and Adaptive Channel Decoding in Faster-than-Nyquist Receiver

On one hand, aiming at the prominent problem of improving the reliability of Faster-than-Nyquist (FTN) wireless transmission, abandoning the channel estimation, we propose the Radial Basis Function Least Squares Support Vector Machine (RBF LSSVM) algorithm based on Intelligent Signal Processing for FTN equalization and form FTN time domain transverse filter equalizer. The model of FTN wireless communication system based on equalization of LSSVM algorithm is established by adding 50-bit training sample sequence module. On the other hand, in the FTN transmission BPSK modulation system receiving terminal, we propose joint research equalization of LSSVM algorithm and adaptive channel decoding scheme of Chase algorithm, to improve the reliable transmission performance of FTN wireless communication. The threshold value is 20, and the adaptive 2-D Turbo Product Codes (TPC) encoding and decoding is simulated by four iterations. The BER performance of FTN wireless communication combined with LSSVM algorithm and TPC adaptive decoding is simulated.

Multimedia Archives: New Digital Filters to Correct Equalization Errors on Digitized Audio Tapes

Multimedia archives face the problem of obsolescing and degrading analogue media (e.g., speech and music recordings and video art). In response, researchers in the field have recently begun studying ad hoc tools for the preservation and access of historical analogue documents. This paper investigates the active preservation process of audio tape recordings, specifically focusing on possible means for compensating equalization errors introduced in the digitization process. If the accuracy of corrective equalization filters is validated, an archivist or musicologist would be able to experience the audio as a historically authentic document such that their listening experience would not require the recovery of the original analogue audio document or the redigitization of the audio. Thus, we conducted a MUSHRA-inspired perception test (n = 14) containing 6 excerpts of electronic music (3 stimuli recorded NAB and 3 recorded CCIR). Participants listened to 6 different equalization filters for each stimulus and rated them in terms of similarity. Filters included a correctly digitized “Reference,” an intentionally incorrect “Foil” filter, and a subsequent digital correction of the Foil filter that was produced with a MATLAB script. When stimuli were collapsed according to their filter type (NAB or CCIR), no significant differences were observed between the Reference and MATLAB correction filters. As such, the digital correction appears to be a promising method for compensation of equalization errors although future study is recommended, specifically containing an increased sample size and additional correction filters for comparison.

An improved LMS algorithm for mode demultiplexing in frequency domain

In order to solve the problem that the traditional frequency domain least mean square (FD-LMS) algorithm will lose efficacy with the increase of differential mode group delay (DMGD) when the algorithm is used for demultiplexing of the 6×6 mode division multiplexing (MDM) system, an improved FD-LMS demultiplexing algorithm is proposed. By improving the error signal calculation method, the convergence performance of the output signal of the equalization filter is improved, and the steady-state error of the algorithm is reduced. Besides, the equalization performance of the traditional FD-LMS algorithm is compared with the improved FD-LMS algorithm. Simulation results show that the improved FD-LMS algorithm has great advantage over the traditional FD-LMS algorithm in demultiplexing performance on the premise that the computation complexity does not significantly increase. The optical signal to noise ratio (OSNR) penalty of the improved FD-LMS algorithm is 2.6 dB lower than that of traditional FD-LMS algorithm at a transmission distance of 80 km with DMGD is 50 ps/km.

Light microscopic iris classification using ensemble multi-class support vector machine.

Similar to other biometric systems such as fingerprint, face, DNA, iris classification could assist law enforcement agencies in identifying humans. Iris classification technology helps law-enforcement agencies to recognize humans by matching their iris with iris data sets. However, iris classification is challenging in the real environment due to its invertible and complex texture variations in the human iris. Accordingly, this article presents an improved Oriented FAST and Rotated BRIEF with Bag-of-Words model to extract distinct and robust features from the iris image, followed by ensemble multi-class-SVM to classify iris. The proposed methodology consists of four main steps; first, iris image normalization and enhancement; second, localizing iris region; third, iris feature extraction; finally, iris classification using ensemble multi-class support vector machine. For preprocessing of input images, histogram equalization, Gaussian mask and median filters are applied. The proposed technique is tested on two benchmark databases, that is, CASIA-v1 and iris image database, and achieved higher accuracy than other existing techniques reported in state of the art.

COVID-DeepNet: Hybrid Multimodal Deep Learning System for Improving COVID-19 Pneumonia Detection in Chest X-ray Images

Coronavirus (COVID-19) epidemic outbreak has devastating effects on daily lives and healthcare systems worldwide. This newly recognized virus is highly transmissible, and no clinically approved vaccine or antiviral medicine is currently available. Early diagnosis of infected patients through effective screening is needed to control the rapid spread of this virus. Chest radiography imaging is an effective diagnosis tool for COVID-19 virus and follow-up. Here, a novel hybrid multimodal deep learning system for identifying COVID-19 virus in chest X-ray (CX-R) images is developed and termed as the COVID-DeepNet system to aid expert radiologists in rapid and accurate image interpretation. First, Contrast-Limited Adaptive Histogram Equalization (CLAHE) and Butterworth bandpass filter were applied to enhance the contrast and eliminate the noise in CX-R images, respectively. Results from two different deep learning approaches based on the incorporation of a deep belief network and a convolutional deep belief network trained from scratch using a large-scale dataset were then fused. Parallel architecture, which provides radiologists a high degree of confidence to distinguish healthy and COVID-19 infected people, was considered. The proposed COVID-DeepNet system can correctly and accurately diagnose patients with COVID-19 with a detection accuracy rate of 99.93%, sensitivity of 99.90%, specificity of 100%, precision of 100%, F1-score of 99.93%, MSE of 0.021%, and RMSE of 0.016% in a large-scale dataset. This system shows efficiency and accuracy and can be used in a real clinical center for the early diagnosis of COVID-19 virus and treatment follow-up with less than 3 s per image to make the final decision.

COVID-CheXNet: hybrid deep learning framework for identifying COVID-19 virus in chest X-rays images.

The outbreaks of Coronavirus (COVID-19) epidemic have increased the pressure on healthcare and medical systems worldwide. The timely diagnosis of infected patients is a critical step to limit the spread of the COVID-19 epidemic. The chest radiography imaging has shown to be an effective screening technique in diagnosing the COVID-19 epidemic. To reduce the pressure on radiologists and control of the epidemic, fast and accurate a hybrid deep learning framework for diagnosing COVID-19 virus in chest X-ray images is developed and termed as the COVID-CheXNet system. First, the contrast of the X-ray image was enhanced and the noise level was reduced using the contrast-limited adaptive histogram equalization and Butterworth bandpass filter, respectively. This was followed by fusing the results obtained from two different pre-trained deep learning models based on the incorporation of a ResNet34 and high-resolution network model trained using a large-scale dataset. Herein, the parallel architecture was considered, which provides radiologists with a high degree of confidence to discriminate between the healthy and COVID-19 infected people. The proposed COVID-CheXNet system has managed to correctly and accurately diagnose the COVID-19 patients with a detection accuracy rate of 99.99%, sensitivity of 99.98%, specificity of 100%, precision of 100%, F1-score of 99.99%, MSE of 0.011%, and RMSE of 0.012% using the weighted sum rule at the score-level. The efficiency and usefulness of the proposed COVID-CheXNet system are established along with the possibility of using it in real clinical centers for fast diagnosis and treatment supplement, with less than 2 s per image to get the prediction result.

Optical Signal Processing Aided by Optical Frequency Combs

This paper highlights certain optical signal processing (OSP) functions that can be achieved when using optical frequency combs. An optical frequency comb can provide many narrow-linewidth, mutually coherent, and equidistant optical carriers, making it a potentially useful tool for facilitating efficient signal processing functions. In this paper, we explore the following general topics: (a) tailoring a comb for enhanced usability in OSP, such as inserting more lines, generating Nyquist pulse trains, and regenerating a comb at a distance; (b) achieving a tunable linear and nonlinear filter for correlation, equalization, and Volterra filtering, and (c) enabling various transmission and networking functions, such as multicasting, format conversion, and dynamic bandwidth allocation.

A subband implementation of a multichannel and multiple position adaptive room response equalizer

This paper deals with a subband implementation of an adaptive multichannel and multiple position room response equalization system capable of improving the listening experience in a real environment. The main focus of the approach is the introduction of a subband identification procedure capable of estimating in real-time the magnitude responses of a system with a great accuracy and a quick convergence. These aspects are very important also for the equalization procedure to obtain a good accuracy and quick development of the equalization filters as function of the environment variations. Several experiments have been performed in a real scenario to test the performance of the algorithm and to show the beneficial effects of the subband identification on the equalization procedure.

String instrument acoustic transfer processing

A real-time method of string instrument acoustic transfer which includes damping is proposed. Acoustic transfer of string instruments is relevant when trying to make a non-resonant instrument, such as an electric guitar, sound more similar to an acoustic guitar. Unlike previous acoustic transfer methods which only perform equalization, this method includes damping changes by using a time-varying filter which adds frequency-dependent exponential damping. Efficient digital filters are fit to bridge admittance measurements of an acoustic guitar and used to create equalization filters as well as damping correction filters. The damping correction filters work in real-time as they are triggered by onset and pitch detection of the signal measured through an under saddle pickup to determine the intensity of the damping.

Perceptual Evaluation of Mitigation Approaches of Impairments due to Spatial Undersampling in Binaural Rendering of Spherical Microphone Array Data

Spherical microphone arrays (SMAs) are widely used to capture spatial sound fields that can then be rendered in various ways as a virtual acoustic environment (VAE) including headphone-based binaural synthesis. Several practical limitations have a significant impact on the fidelity of the rendered VAE. The finite number of microphones of SMAs leads to spatial undersampling of the captured sound field, which, on the one hand, induces spatial aliasing artifacts and, on the other hand, limits the order of the spherical harmonics (SH) representation. Several approaches have been presented in the literature that aim to mitigate the perceptual impairments due to these limitations. In this article, we present a listening experiment evaluating the perceptual improvements of binaural rendering of undersampled SMA data that can be achieved using state-of-the-art mitigation approaches. In particular, we examined the Magnitude Least-Squares algorithm, the Bandwidth Extraction Algorithm for Microphone Arrays, Spherical Head Filters, SH Tapering, and a newly proposed equalization filter. In the experiment, subjects rated the perceived differences between a dummy head and the corresponding SMA auralization. We found that most mitigation approaches lead to significant perceptual improvements, even though audible differences to the reference remain.

Computer-aided diabetic retinopathy diagnostic model using optimal thresholding merged with neural network

PurposeDiabetic retinopathy (DR) is a central root of blindness all over the world. Though DR is tough to diagnose in starting stages, and the detection procedure might be time-consuming even for qualified experts. Nowadays, intelligent disease detection techniques are extremely acceptable for progress analysis and recognition of various diseases. Therefore, a computer-aided diagnosis scheme based on intelligent learning approaches is intended to propose for diagnosing DR effectively using a benchmark dataset.Design/methodology/approachThe proposed DR diagnostic procedure involves four main steps: (1) image pre-processing, (2) blood vessel segmentation, (3) feature extraction, and (4) classification. Initially, the retinal fundus image is taken for pre-processing with the help of Contrast Limited Adaptive Histogram Equalization (CLAHE) and average filter. In the next step, the blood vessel segmentation is carried out using a segmentation process with optimized gray-level thresholding. Once the blood vessels are extracted, feature extraction is done, using Local Binary Pattern (LBP), Texture Energy Measurement (TEM based on Laws of Texture Energy), and two entropy computations – Shanon's entropy, and Kapur's entropy. These collected features are subjected to a classifier called Neural Network (NN) with an optimized training algorithm. Both the gray-level thresholding and NN is enhanced by the Modified Levy Updated-Dragonfly Algorithm (MLU-DA), which operates to maximize the segmentation accuracy and to reduce the error difference between the predicted and actual outcomes of the NN. Finally, this classification error can correctly prove the efficiency of the proposed DR detection model.FindingsThe overall accuracy of the proposed MLU-DA was 16.6% superior to conventional classifiers, and the precision of the developed MLU-DA was 22% better than LM-NN, 16.6% better than PSO-NN, GWO-NN, and DA-NN. Finally, it is concluded that the implemented MLU-DA outperformed state-of-the-art algorithms in detecting DR.Originality/valueThis paper adopts the latest optimization algorithm called MLU-DA-Neural Network with optimal gray-level thresholding for detecting diabetic retinopathy disease. This is the first work utilizes MLU-DA-based Neural Network for computer-aided Diabetic Retinopathy diagnosis.

Analog I/Q FIR Filter in 55-nm SiGe BiCMOS for 16-QAM Optical Communications at 112 Gb/s

We propose a novel implementation of a complex analog equalization filter for the compensation of frequency-dependent variations in coherent optical links. The analog compensation filter can be used in coherent-lite optical communication links where digital signal processing (DSP) is removed to limit the complexity and power consumption. In these links, the filter can compensate for electrical bandwidth limitations and distortion introduced by chromatic dispersion in the fiber. The complex filter is implemented by combining four distributed analog finite-impulse response (FIR) filters to obtain the necessary response. The filter delays are implemented using active delay cell structures to create a compact solution. The analog filter is implemented in a 55-nm BiCMOS technology and consumes 185-mW core power for five complex filter taps. Performance is evaluated using the S-parameter measurements, noise and linearity measurements, and real-time system experiments using 112-Gb/s 16-QAM-modulated signals.

Increased Mammogram Image Contrast Using Histogram Equalization And Gaussian In The Classification Of Breast Cancer

Breast cancer is one of the most common diseases among women in several countries. One of the most common methods to diagnose breast cancer is mammography. In this study, we propose a classification study to differentiate benign and malignant breast tumors based on mammogram image. The proposed system includes five major steps, i.e. preprocessing, histogram equalization, convolution, feature extraction, and classification. Image is cropped using region of interest (ROI) at preprocessing stage. In this study, we perform image contrast quality enhancement of the mammogram to view the breast cancer better. Image contrast enhancement uses histogram equalization and Gaussian filter. Gray-Level Co-Occurrence Matrix (GLCM) is used to extract the mammogram features. There are five features used i.e. entropy, correlation, contrast, homogeneity, and variance. The last step is to classify using naïve Bayes classifier (NBC) and k-nearest neighbor (KNN). Based on the hypothesis, the accuracy of NBC method is 90% and the accuracy of KKN method is 87.5%. So, the mammogram image contrast enhancement is well performed.

A 12.8-Gbaud ADC-Based Wireline Receiver With Embedded IIR Equalizer

This article demonstrates an analog-to-digital converter (ADC)-based receiver for NRZ/PAM4 modulation, featuring a time-to-digital converter (TDC)-assisted multi-bit/cycle asynchronous successive approximation register (SAR) ADC with embedded IIR equalization filter driven by the differential source followers with an active gain. It re-uses the existing sampling network of time-interleaved (TI) ADCs and incorporates active Gm-C integrators to form a tunable IIR equalizer response. The prototype is fabricated in 65-nm complementary metal–oxide–semiconductor (CMOS) and achieves an efficiency of 2.43-pJ/b using the 12.8-Gbuad PAM4 modulation scheme. The eight-way TI ADC measures 4.84 peak effective number of bit with power consumption of 36.3 mW while occupying 0.24 mm2 core area.

On the Effects of Channel Sparsity on Joint Estimators in Aeronautical Telemetry

Minimum mean-squared error (MMSE) equalizers are a viable solution to mitigate the frequency selectivity of the aeronautical telemetry channel. Because the MMSE equalizer filter coefficients are a function of the carrier frequency offset (CFO), the equivalent discrete-time channel, and the noise variance, reliable estimates of those parameters are required. The CFO is due primarily to the high velocity of the airborne transmitter. Because the equivalent discrete-time channel in aeronautical telemetry is sparse, the performance of the joint estimator based on a sparse channel estimator is superior to the performance of a traditional nonsparse maximum-likelihood-inspired joint estimator. The improvement is seen in lower estimator error variances for the parameters, particularly for the CFO and channel estimate, and in the postequalizer bit-error rate. Simulation results demonstrate that the postequalizer bit-error rate using the sparse estimator is almost as good as that using ideal estimators.

DEHAZING RESEARCH ON BRIGHTNESS EQUALIZATION MODEL OF DRONE IMAGE

Abstract. Due to the rapid development of drone technology, aerial imagery of drones is increasingly used in various fields. However, the aerial image of the drone is highly susceptible to weather conditions during the imaging process. Most aerial images are inevitably affected by fog when they are acquired. Due to the scattering and absorption of the atmosphere, the aerial image of the drone in foggy days has the characteristics of low contrast and unclear scenery. Due to the scattering and absorption of the atmosphere, the aerial image of drone acquired in the foggy environment has the characteristics of low contrast and unclear scenery. The Defogging technology for aerial image of drone can obtain a large amount of useful information in a pictures with low information amount through a certain image processing method, and convert the image with low information amount into a useful image. Therefore, the image processing research carried out for such image degradation caused by natural phenomena has universal practical significance. Aiming at the problem that the aerial image of drone is often affected by haze and the image is blurred and the image quality is degraded, this paper proposes a new model for defogging aerial image of drone. The brightness equalization model is used to improve the degraded image with fog defects. The brightness equalization model obtains the brightness channel of the original image based on the HSI transform. The brightness equalization filter is used to dynamically adjust the brightness to the appropriate interval to achieve the purpose of defogging and then further optimizes the result image by using Gaussian blur and color reshaping. Two images with fog problems were compared, using the brightness equalization model of this paper. And the quality evaluation parameters are selected to evaluate the processing results of the dehazing model. The average value of the images processed by the model is more suitable and the main quality evaluation parameters such as standard deviation and entropy are better than those of the original image.The experimental results show that the brightness equalization model can effectively remove the influence of fog in the aerial image of the drone and improve the visual effect of the image.