Adaptive Median Filtering Algorithm Based on Divide and Conquer and Its Application in CAPTCHA Recognition

As the first barrier to protect cyberspace,the CAPTCHA has made significant contributions to maintaining Internet security and preventing malicious attacks.By researching the CAPTCHA,we can find its vulnerability and improve the security of CAPTCHA.Recently,many studies have shown that improving the image preprocessing effect of the CAPTCHA,which can achieve a better recognition rate by the state-of-theart machine learning algorithms.There are many kinds of noise and distortion in the CAPTCHA images of this experiment.We propose an adaptive median filtering algorithm based on divide and conquer in this paper.Firstly,the filtering window data quickly sorted by the data correlation,which can greatly improve the filtering efficiency.Secondly,the size of the filtering window is adaptively adjusted according to the noise density.As demonstrated in the experimental results,the proposed scheme can achieve superior performance compared with the conventional median filter.The algorithm can not only effectively detect the noise and remove it,but also has a good effect in preservation details.Therefore,this algorithm can be one of the most strong tools for various CAPTCHA image recognition and related applications.

Median Filtering Forensics Scheme for Color Images Based on Quaternion Magnitude-Phase CNN

In the paper,a convolutional neural network based on quaternion transformation is proposed to detect median filtering for color images.Compared with conventional convolutional neural network,color images can be processed in a holistic manner in the proposed scheme,which makes full use of the correlation between RGB channels.And due to the use of convolutional neural network,it can effectively avoid the one-sidedness of artificial features.Experimental results have shown the scheme’s improvement over the state-of-the-art scheme on the accuracy of color image median filtering detection.

Median Filtering Detection Based on Quaternion Convolutional Neural Network

Median filtering is a nonlinear signal processing technique and has an advantage in the field of image anti-forensics.Therefore,more attention has been paid to the forensics research of median filtering.In this paper,a median filtering forensics method based on quaternion convolutional neural network(QCNN)is proposed.The median filtering residuals(MFR)are used to preprocess the images.Then the output of MFR is expanded to four channels and used as the input of QCNN.In QCNN,quaternion convolution is designed that can better mix the information of different channels than traditional methods.The quaternion pooling layer is designed to evaluate the result of quaternion convolution.QCNN is proposed to features well combine the three-channel information of color image and fully extract forensics features.Experiments show that the proposed method has higher accuracy and shorter training time than the traditional convolutional neural network with the same convolution depth.

Wavelet domain denoising method based on multistage median filtering

There are two main problems in the threshold denoising method based on wavelet transform. One is the difficulty of threshold selection, and the other is the inconsistence of the dip and curved events in the low signal-to-noise ratio （SNR） seismic data after denoising. In image denoising, multistage median filtering can preserve the details of the signal. So we proposed a denoising algorithm in wavelet transform domain based on multistage median filtering. Using this method the flat region and the edge region are differentiated by the difference between the maximum mid-value and the minimum mid-value, which preserves the details, thus improves the denoising effect. The simulation data and the real data processing results reveal that this method has stronger ability in separating signal from noise than that of the threshold denoising method.

Recognition of Partially Occluded Objects Based on Segmentation by Median Filtering of Extended Direction Code

The image contour is segmented into lines, arcs and smooth curves by median filtering of extended direction code. Based on this segmentation, a set of new local invariant features are proposed to recognize partially occluded objects, which is more reasonable compared with conventional corner features. The matching results of some typical examples shows that these features are robust ,effective in recognition.

Real Time Speed Bump Detection Using Gaussian Filtering and Connected Component Approach

An Intelligent Transportation System (ITS) is a new system developed for the betterment of user in traffic and transport management domain area for smart and safe driving. ITS subsystems are Emergency vehicle notification systems, Automatic road enforcement, Collision avoidance systems, Automatic parking, Map database management, etc. Advance Driver Assists System (ADAS) belongs to ITS which provides alert or warning or information to the user during driving. The proposed method uses Gaussian filtering and Median filtering to remove noise in the image. Subsequently image subtraction is achieved by subtracting Median filtered image from Gaussian filtered image. The resultant image is converted to binary image and the regions are analyzed using connected component approach. The prior work on speed bump detection is achieved using sensors which are failed to detect speed bumps that are constructed with small height and the detection rate is affected due to erroneous identification. And the smartphone and accelerometer methodologies are not perfectly suitable for real time scenario due to GPS error, network overload, real-time delay, accuracy and battery running out. The proposed system goes very well for the roads which are constructed with proper painting irrespective of their dimension.

An Adaptive Median Filter Algorithm Based on B-spline Function

According to the B-spline convolution mask, first, the contrast sensitiveness （CS） is computed and then is viewed as a noise sensitiveness coeficient （NSC） to adaptively determine a noise-recognized threshold value. Based on the noise density function （NDF） in a 3×3 window, the filtering window size is adaptively adjusted, and then a median filter is used to eliminate the noise-marked pixels. The experiment results show that the proposed algorithm can preserve image detail information well and effectively remove the noises, particularly the impulse noises that is also called salt-and-pepper noises superimposed on the computed tomography （CT） and magnetic resonance imaging （MRI） medical images.

Adaptive Threshold Median Filter for Multiple-Impulse Noise

Attenuating the noises plays an essential role in the image processing. Almost all the traditional median filters concern the removal of impulse noise having a single layer, whose noise gray level value is constant. In this paper, a new adaptive median filter is proposed to handle those images corrupted not only by single layer noise. The adaptive threshold median filter （ATMF） has been developed by combining the adaptive median filter （AMF） and two dynamic thresholds. Because of the dynamic threshold being used, the ATMF is able to balance the removal of the multiple-impulse noise and the quality of image. Comparison of the proposed method with traditional median filters is provided. Some visual examples are given to demonstrate the performance of the proposed filter.

Filtering algorithm of line structured light for long-distance obstacle detection

Since unmanned ground vehicles often encounter concave and convex obstacles in wild ground, a filtering algorithm using line structured light to detect these long distance obstacles is proposed. For the line structured light image, a ranked-order based adaptively extremum median （RAEM） filter algorithm on salt and pepper noise is presented. In the algorithm, firstly effective points and noise points in a filtering window are differentiated; then the gray values of noise points are replaced by the medium of gray values of the effective pixels, with the efficient points＇ gray values unchanged; in the end this algorithm is proved to be efficient by experiments. Experimental resuits demonstrate that the image blur, resulting into proposed algorithm can remove noise points effectively and minimize the protecting the edge information as much as possible.

Some Problems of 2D Morphological and Median Filters

In this paper we mainly discussed some problems of 2D morpnological and median filters .The differences between 1D and 2D morphological and median filters arc also described. It can be seen that many propcrties of ID finers arc invalid for 2D filters. Som cxamples and cxpcriments are gived to show these problems.

Classification of Multi-view Digital Mammogram Images Using SMO-WkNN

Breast cancer(BCa)is a leading cause of death in the female population across the globe.Approximately 2.3 million new BCa cases are recorded globally in females,overtaking lung cancer as the most prevalent form of cancer to be diagnosed.However,the mortality rates for cervical and BCa are significantly higher in developing nations than in developed countries.Early diagnosis is the only option to minimize the risks of BCa.Deep learning(DL)-based models have performed well in image processing in recent years,particularly convolutional neural network(CNN).Hence,this research proposes a DL-based CNN model to diagnose BCa from digitized mammogram images.The main objective of this research is to develop an accurate and efficient early diagnosis model for BCa detection.This proposed model is a multi-view-based computer-aided diagnosis(CAD)model,which performs the diagnosis of BCa on multi-views of mammogram images like medio-lateral-oblique(MLO)and cranio-caudal(CC).The digital mammogram images are collected from the digital database for screening mammography(DDSM)dataset.In preprocessing,median filter and contrast limited adaptive histogram equalization(CLAHE)techniques are utilized for image enhancement.After preprocessing,the segmentation is performed using the region growing(RG)algorithm.The feature extraction is carried out from the segmented images using a pyramidal histogram of oriented gradients(PHOG)and the AlextNet model.Finally,the classification is performed using the weighted k-nearest neighbor(WkNN)optimized with sequential minimal optimization(SMO).The classified images are evaluated based on accuracy,recall,precision,specificity,f1-score,and mathews correlation coefficient(MCC).Additionally,the false positive and error rates are evaluated.The proposed model obtained 98.57%accuracy,98.61%recall,99.25%specificity,98.63%precision,97.93%f1-score,96.26%MCC,0.0143 error rate,and 0.0075 false positive rate(FPR).Compared to the existing models,the research model has obtained better performances and outperformed the other models.

De-Noising Brain MRI Images by Mixing Concatenation and Residual Learning(MCR)

Brain magnetic resonance images(MRI)are used to diagnose the different diseases of the brain,such as swelling and tumor detection.The quality of the brain MR images is degraded by different noises,usually salt&pepper and Gaussian noises,which are added to the MR images during the acquisition process.In the presence of these noises,medical experts are facing problems in diagnosing diseases from noisy brain MR images.Therefore,we have proposed a de-noising method by mixing concatenation,and residual deep learning techniques called the MCR de-noising method.Our proposed MCR method is to eliminate salt&pepper and gaussian noises as much as possible from the brain MRI images.The MCR method has been trained and tested on the noise quantity levels 2%to 20%for both salt&pepper and gaussian noise.The experiments have been done on publically available brain MRI image datasets,which can easily be accessible in the experiments and result section.The Structure Similarity Index Measure(SSIM)and Peak Signal-to-Noise Ratio(PSNR)calculate the similarity score between the denoised images by the proposed MCR method and the original clean images.Also,the Mean Squared Error(MSE)measures the error or difference between generated denoised and the original images.The proposed MCR denoising method has a 0.9763 SSIM score,84.3182 PSNR,and 0.0004 MSE for salt&pepper noise;similarly,0.7402 SSIM score,72.7601 PSNR,and 0.0041 MSE for Gaussian noise at the highest level of 20%noise.In the end,we have compared the MCR method with the state-of-the-art de-noising filters such as median and wiener de-noising filters.

Analytic Beta-Wavelet Transform-Based Digital Image Watermarking for Secure Transmission

The rapid development in the information technology field has introduced digital watermark technologies as a solution to prevent unauthorized copying and redistribution of data.This article introduces a self-embedded image verification and integrity scheme.The images are firstly split into dedicated segments of the same block sizes.Then,different Analytic Beta-Wavelet(ABW)orthogonal filters are utilized for embedding a self-segment watermark for image segment using a predefined method.ABW orthogonal filter coefficients are estimated to improve image reconstruction under different block sizes.We conduct a comparative study comparing the watermarked images using three kinds of ABW filters for block sizes 64×64,128×128,and 256×256.We embed the watermark using the ABW-based image watermarking method in the 2-level middle frequency sub-bands of the ABW digital image coefficients.The imperceptibility and robustness of the ABW-based image watermarking method image is evaluated based on the Peak Signal to Noise Ratio(PSNR)and Correlation coefficient values.From the implementation results,we came to know that this ABW-based image watermarking method can withstand many image manipulations compared to other existing methods.

Optimal Deep Learning Based Inception Model for Cervical Cancer Diagnosis

Prevention of cervical cancer becomes essential and is carried out by the use of Pap smear images.Pap smear test analysis is laborious and tiresome work performed visually using a cytopathologist.Therefore,automated cervical cancer diagnosis using automated methods are necessary.This paper designs an optimal deep learning based Inception model for cervical cancer diagnosis(ODLIM-CCD)using pap smear images.The proposed ODLIM-CCD technique incorporates median filtering(MF)based pre-processing to discard the noise and Otsu model based segmentation process.Besides,deep convolutional neural network(DCNN)based Inception with Residual Network(ResNet)v2 model is utilized for deriving the feature vectors.Moreover,swallow swarm optimization(SSO)based hyperparameter tuning process is carried out for the optimal selection of hyperparameters.Finally,recurrent neural network(RNN)based classification process is done to determine the presence of cervical cancer or not.In order to showcase the improved diagnostic performance of the ODLIM-CCD technique,a series of simulations occur on benchmark test images and the outcomes highlighted the improved performance over the recent approaches with a superior accuracy of 0.9661.

Path Recognition Algorithm of Tracking Robots Based on Image Sensor

Path recognition is an inevitable core technology in the development of tracking robot. In this paper,the path tracking system of tracking robot can be realized by image sensor module based on camera to obtain lane image information,and then extract the path through visual servo. The whole system can be divided into seven modules: micro control unit( MCU) processor module,image acquisition module,debugging module,motor drive module,servo drive module,speed sensor module,and voltage conversion module.In image pre-processing part,there is an introduction of binarization processing and the median filtering to strengthen the image information. About recognition algorithm,three key variables which are changed in the movement state are discussed and there are also many auxiliary algorithms that help to improve the path recognition.The experiment can verify that the whole system can accurately abstract the black guide lines from the white track and make the robot moving fast and stable by following the road parameters and conditions.

Multi-level denoising and enhancement method based on wavelet transform for mine monitoring

Based on low illumination and a large number of mixed noises contained in coal mine, denoising with one method usually cannot achieve good results, so a multi-level image denoising method based on wavelet correlation relevant inter-scale is presented. Firstly, we used directional median filter to effectively reduce impulse noise in the spatial domain, which is the main cause of noise in mine. Secondly, we used a Wiener filtration method to mainly reduce the Gaussian noise, and then finally used a multi-wavelet transform to minimize the remaining noise of low-light images in the transform domain. This multi-level image noise reduction method combines spatial and transform domain denoising to enhance benefits, and effectively reduce impulse noise and Gaussian noise in a coal mine, while retaining good detailed image characteristics of the underground for improving quality of images with mixing noise and effective low-light environment.

DENOISING METHODS IN INTERFEROMETRIC SAR IMAGES PROCESSING

Thermal and “Speckle” noise is an obstacle to generate the digital elevation model(DEM) from interferogram by 2 D SAR images. According to the theory of interferometry, the main sources of noise in interferogram are discussed. On the basis of the character of noise in the interferogram, the low pass filter, median filter and wavelet transform are investigated. Wavelet transform is forwarded as the most effective method to eliminate the InSAR noise because it can keep the resolution of the images during eliminating the noise. The raw data verify the validity and effectiveness of wavelet transform.

Automatic Segmentation and Detection System for Varicocele Using Ultrasound Images

The enlarged veins in the pampiniform venous plexus,known as varicocele disease,are typically identified using ultrasound scans.Themedical diagnosis of varicocele is based on examinations made in three positions taken to the right and left testicles of the male patient.The proposed system is designed to determine whether a patient is affected.Varicocele is more frequent on the left side of the scrotum than on the right and physicians commonly depend on the supine position more than other positions.Therefore,the experimental results of this study focused on images taken in the supine position of the left testicles of patients.There are two possible vein structures in each image:a cross-section(circular)and a tube(non-circular)structure.This proposed system identifies dilated(varicocele)veins of these structures in ultrasound images in three stages:preprocessing,processing,and detection and measurement.These three stages are applied in three different color modes:Grayscale,Red-Green-Blue(RGB),and Hue,Saturation,and Value(HSV).In the preprocessing stage,the region of interest enclosing the pampiniform plexus area is extracted using a median filter and threshold segmentation.Then,the processing stage employs different filters to perform image denoising.Finally,edge detection is applied in multiple steps and the detected veins are measured to determine if dilated veins exist.Overall implementation results showed the proposed system is faster andmore effective than the previous work.

Image Filter for Removal of Impulse Noise Based on Directional Derivatives

One of the most common image processing tasks involves the removal of noise from images. Noise can be introduced during image capture, during transmission, or during storage. For design purposes, noise sources are frequently approximated by random variables with a known probability distribution. One common noise model corrupts a signal by introducing impulses. And the surface of the image disturbed by impulse noise displays many peaks or vales. According to the characteristic of impulse noise, a novel algorithm is proposed to the detection of impulse noise point from images based on directional derivatives. First, the theory of calculus on directional derivatives is introduced in detail. Then it is applied to the field of image to removing noise with the discrete form derived from its continuous mathematical model. And a number of contrasting simulations illustrate that our algorithm not only can preserve the structure information while removing impulse noise but also can mostly save the gray value of the pixels undisturbed by noise. In addition, the comparisons of the filtering performance for removing impulse noise are analyzed in detail in the case of different noise densities, and also show that the algorithm suggested outperforms the conventional filter algorithms such as mean filter, median filter and so on in speed and impulse noise reduction, especially in random-valued impulse noise reduction. So it is a very good alternative to the existing schemes.

An enhanced PSO-DEFS based feature selection with biometric authentication for identi¯cation of diabetic retinopathy

Recently,automatic diagnosis of diabetic retinopathy(DR)from the retinal image is the most significant ressearch topic in the medical applications.Diabetic macular edema(DME)is the.major reason for the loss of vision in patients suffering fom DR.Early identification of the DR enables to prevent the vision loss and encourage diabetic control activities.Many techniques are.developed to diagnose the DR.The major drawbacks of the existing techniques are low accuracy and high time complexity.To owercome these issues,this paper propases an enhanced particle swarm optimization differential evolution feature selection(PSO DEFS)based feature selection approach with biometric aut hentication for the identification of DR.Initially,a hybrid median filter(HMF)is used for pre processing the input images.Then,the pre-processed images are embedded with each other by using least significant bit(LSB)for authentication purpose.Si-multaneously,the image features are extracted using convoluted local tetra pattern(CLTrP)and Tamura features.Feature selection is performed using PSO DEFS and PSO-gravitational search algorithm(PSO GSA)to reduce time complexity.Based on some performance metrics,the PSO-DEFS is chosen as a better choice for feature selection.The feature selection is performed based on the fitness value.A multi-relevance vector machine(M-RVM)is introduced to dlassify the 13 normal and 62 abnormal images among 75 images from 60 patients.Finally,the DR patients are further dassified by M-RVM.The experimental results exhibit that the proposed approach achieves better accuracy,sensitivity,and specificity than the exist ing techniques.