Salt-pepper Noise Research Articles

Advanced Dual-Input Artificial Optical Synapse for Recognition and Generative Neural Network

Perovskite materials have emerged as leading candidates for optical synaptic devices due to their superior photosensitivity and tunable optoelectronic properties. However, the practical application of perovskite-based optoelectronic synaptic transistors has been hindered by issues of poor stability and high toxicity. This study developed an artificial synaptic thin film transistor (TFT) based on a Cs2AgBiBr6/InOx bilayer. The device demonstrated synaptic behavior under optoelectronic hybrid stimulation (largest wavelength ~520nm), such as excitatory postsynaptic current (EPSC), inhibitory postsynaptic current (IPSC), paired-pulse facilitation (PPF), spike-timing-dependent plasticity (STDP), short-term memory (STM) and long-term memory (LTM). Notably, the “light writing and voltage erasing” characteristics of these devices could be utilized to construct a convolutional neural network (CNN) classifier for the CIFAR-10 dataset, demonstrating noise tolerance close to the human eye. The loss in recognition accuracy was within 1% when Gaussian white noise and salt pepper noise were added. Furthermore, these devices exhibited great potential in Generative Adversarial Networks (CycleGAN), with the generated image quality achieving levels of 0.637 and 0.715 in Improved Perceptual Image Processing System (IPIPS) and Structural Similarity Index (SSIM) evaluation metrics for the zebra dataset, indicating good image quality. This study indicates that our Cs2AgBiBr6-based artificial optical synaptic thin-film transistors (TFTs) are promising for sensing and in-memory computing applications.

Multi-objective optimization for vector quantization via genetic algorithm

The issue of image compression continues to be a subject of ongoing research within the domain of image processing, particularly in the context of medical applications. The quality of the decompressed image can vary depending on the accuracy of the compression technique, resulting in either fine or distorted details. Therefore, the diagnostic procedure performed by medical professionals is contingent upon the precision of the compression and decompression process. In addition, the compression of medical images serves to decrease the amount of storage required, thereby enabling faster transmission over computer networks through the reduction of their bit size. This paper proposes a hybrid mix of the discrete wavelet transform (DWT) technique and vector quantization (VQ) to improve the compression technique of medical images. The aim of the proposed compression technique is to preserve diagnostic image information while achieving a high compression ratio. First, noise in medical images is caused by salt pepper noise. At the same time, the edges of the images are maintained in sharpness and context. Then, a lossless compression method is applied to the wavelet coefficients of the subband with the lowest frequency, while the thresholding method was used to efficiently construct coefficients for high-frequency sub-bands. This process will produce a traditional VQ, which is estimated via the Genetic Algorithm (GA) with fuzzy clustering. While Arithmetic encoded theory was being utilized to quantize coefficients, the proposed compression technique was evaluated by dividing the image into two levels and three levels of sub-bands, respectively, in two different scenarios. Comparing the decompressed image to the filtered image by means of different evaluation metrics, the proposed method can enhance compression performance and strike a balance between compression ratio and image visual quality.

Ice-Core Micro-CT Image Segmentation with Dual Stream Spectrum Deconvolution Neural Network and Gaussian Mixture Model

Polar ice sheets, or ice cores, are among the most well-known natural archives that might provide crucial historical details about our planet's previous environment. An important factor in establishing the fundamental characteristics of ice, likes pore close-off, albedo, melt events, is the ice-core microstructure. To engulf these complications Ice-Core Micro-CT Image Segmentation with Dual Stream Spectrum Deconvolution Neural Network and Gaussian Mixture Model (ICMCTS-WSOA-DSSDNN-GMM)is proposed. Initially, micro scale CT images are collect from Alfred Wegener Institute ice-core storage as input. Then,data’s are given to pre-processing. In pre-processing, it enhance image brightness, remove salt pepper noise, crop outer ring (carbon fiber casing) to have only ice particles in image utilizing Federated Neural Collaborative Filtering (FNCF).The pre-processing output is given to segmentation. Here, micro scale CT image is utilized for segmenting high-resolution scans using Gaussian Mixture Model (GMM). After that, the segmented images are given to Dual Stream Spectrum Deconvolution Neural Network optimized with Water Strider Optimization Algorithm for classifying micro scale CT images as sintered snow, compacted firn and bubbly ice. The proposed ICMCTS-WSOA-DSSDNN-GMM method is executed in python. The performance of ICMCTS-WSOA-DSSDNN-GMM approach attains 16.24%, 17.90% and 27.7% high accuracy, 14.04%, 25.51% and 19.31% higher precision and 14.36%, 12.65%, 14.51% higher recall analyzed with existing techniques likes Ice-Core Micro-CT Image Segmentation With Deep Learning and Gaussian Mixture Method (ICMCTS-U-net-GMM), Computer Aided Detection of COVID 19 from CT Images Depend on Gaussian Mixture Method with Kernel Support Vector Machines Classifer (ICMCTS-KNN-GMM) and the Gmmseg: Gaussian mixture based generative semantic segmentation methods(ICMCTS-FCN-GMM) respectively.

Complex Noise-Based Phase Retrieval Using Total Variation and Wavelet Transform Regularization

This paper presents a phase retrieval algorithm that incorporates sparsity priors into total variation and framelet regularization. The proposed algorithm exploits the sparsity priors in both the gradient domain and the spatial distribution domain to impose desirable characteristics on the reconstructed image. We utilize structured illuminated patterns in holography, consisting of three light fields. The theoretical and numerical analyses demonstrate that when the illumination pattern parameters are non-integers, the three diffracted data sets are sufficient for image restoration. The proposed model is solved using the alternating direction multiplier method. The numerical experiments confirm the theoretical findings of the lighting mode settings, and the algorithm effectively recovers the object from Gaussian and salt–pepper noise.

Low-cost AI-based solar panel detection drone design and implementation for solar power systems

PurposeThis paper aims to design an AI-based drone that can facilitate the complicated and time-intensive control process for detecting healthy and defective solar panels. Today, the use of solar panels is becoming widespread, and control problems are increasing. Physical control of the solar panels is critical in obtaining electrical power. Controlling solar panel power plants and rooftop panel applications installed in large areas can be difficult and time-consuming. Therefore, this paper designs a system that aims to panel detection.Design/methodology/approachThis paper designed a low-cost AI-based unmanned aerial vehicle to reduce the difficulty of the control process. Convolutional neural network based AI models were developed to classify solar panels as damaged, dusty and normal. Two approaches to the solar panel detection model were adopted: Approach 1 and Approach 2.FindingsThe training was conducted with YOLOv5, YOLOv6 and YOLOv8 models in Approach 1. The best F1 score was 81% at 150 epochs with YOLOv5m. In total, 87% and 89% of the best F1 score and mAP values were obtained with the YOLOv5s model at 100 epochs in Approach 2 as a proposed method. The best models at Approaches 1 and 2 were used with a developed AI-based drone in the real-time test application.Originality/valueThe AI-based low-cost solar panel detection drone was developed with an original data set of 1,100 images. A detailed comparative analysis of YOLOv5, YOLOv6 and YOLOv8 models regarding performance metrics was realized. Gaussian, salt-pepper noise addition and wavelet transform noise removal preprocessing techniques were applied to the created data set under the proposed method. The proposed method demonstrated expressive and remarkable performance in panel detection applications.

잡음 종류와 잡음 제거방식에 따른 영상의 윤곽선 검출 비교

In this paper, we show comparison results of edge detection of images that have additive gaussian noise or salt and pepper noise by using various techniques of noise removal such as filtering, morphology and deep learning based ones. In particular, this present work provides comparison results of noise removal by using gaussian filter, open and close operations of morphology and auto-encoder model followed by carrying out edge detection. Robert cross, Sobel, Prewitt and Canny detectors are used for edge detection of the images with noise removal. Experimental results show that noise removal results are different with characteristics of noise and techniques applied for noise removal. In addition, deep learning based technique, auto-encoder does not always shows superior results of noise removal, particularly in the case of existence of salt-pepper noise. In the experiments, gaussian noise or salt-pepper noise is used and peak signal noise ratio (PSNR) is used for quantitative comparison and the results of edge detection is qualitatively compared from visual perspective.

A Destriping Algorithm for SDGSAT-1 Nighttime Light Images Based on Anomaly Detection and Spectral Similarity Restoration

Remote sensing nighttime lights (NTLs) offers a unique perspective on human activity, and NTL images are widely used in urbanization monitoring, light pollution, and other human-related research. As one of the payloads of sustainable development science Satellite-1 (SDGSAT-1), the Glimmer Imager (GI) provides a new multi-spectral, high-resolution, global coverage of NTL images. However, during the on-orbit testing of SDGSAT-1, a large number of stripes with bad or corrupted pixels were observed in the L1A GI image, which directly affected the accuracy and availability of data applications. Therefore, we propose a novel destriping algorithm based on anomaly detection and spectral similarity restoration (ADSSR) for the GI image. The ADSSR algorithm mainly consists of three parts: pretreatment, stripe detection, and stripe restoration. In the pretreatment, salt-pepper noise is suppressed by setting a minimum area threshold of the connected components. Then, during stripe detections, the valid pixel number sequence and the total pixel value sequence are analyzed to determine the location of stripes, and the abnormal pixels of each stripe are estimated by a clustering algorithm. Finally, a spectral-similarity-based method is adopted to restore all abnormal pixels of each stripe in the stripe restoration. In this paper, the ADSSR algorithm is compared with three representative destriping algorithms, and the robustness of the ADSSR algorithm is tested on different sizes of GI images. The results show that the ADSSR algorithm performs better than three representative destriping algorithms in terms of visual and quantitative indexes and still maintains outstanding performance and robustness in differently sized GI images.

NABILD: Noise And Blur Invariant Local Descriptor for Face Recognition

Most of the existing local descriptors unable to perform well in front of noise and blur variations. Additionally there are very less descriptors persist in literature which are noise and blur invariant. To remedy this challenge the proposed work launch novel descriptor under noise and blur changes so-called Noise and Blur Invariant Local Descriptor (NABILD). With respect to two artificial noises i.e. Gaussian White Noise (GWN) and Salt & Pepper Noise (SPN) with artificial image blurring, the NABILD is introduced. Precisely NABILD takes essentials of two well performed descriptors with respect to noise and blur variations. The first one is Median Robust Extended LBP based on Neighborhood Intensity (MRELBP-NI) and second one is Multiscale Local Phase Quantization (MLPQ). MRELBP-NI is very effective in controlling GWN and SPN due to the capturing of microstructure and macrostructure information. LPQ is the efficient blur invariant descriptor as it quantizes the phase. By considering the merits of both of these descriptors their features are integrated into one framework called as NABILD. To lower down the feature dimension FLDA is deployed and classification is conducted by SVMs. Experiments on ORL face dataset confirm strength of NABILD against other tested descriptors in noise and blur variations. Various literature methods are also outclassed by NABILD.

Correlation-based feature selection using bio-inspired algorithms and optimized KELM classifier for glaucoma diagnosis

Reduced computational time and cost, reduced skilled professional resources, and diagnostic accuracy have made medical diagnosis using computer aided systems (CAD) increasingly popular and can be comfortably employed in the diagnosis of many acute and chronic diseases in ophthalmology, cardiology, cancer detection, etc. There seems to be a growing necessity for the computational algorithms to be robust enough, to identify the abnormality in each of the cases, aiding in early diagnosis. In this paper, a glaucoma diagnostic approach based on the wrapper method employing bio-inspired algorithms, and a Kernel-Extreme Learning Machine (KELM) classifier is proposed. The bio-inspired algorithms are deployed to select feature sub-sets, generating three feature sub-sets from the pre-processed fundus images by adopting a correlation-based feature selection (CFS) approach. The selected features are utilized to train the salp-swarm optimization based KELM, which finds the optimal parameters of the KELM classifier network. The proposed methodology is evaluated on the public and private retinal fundus datasets containing 7280 images. The experimental outcome revealed that the system is able to attain a maximum overall accuracy of 99.61% with 99.89% sensitivity and 100% specificity. A 5-fold cross validation showed 98.78% accuracy ensuring a bias-free classification. Further, by experimenting on degraded images (Gaussian, salt-pepper noise images) of the original dataset, the model achieved extreme robustness with 99.3% accuracy. The proposed method is compared with other similar methods, which showed the efficiency of our method. The framework proposed can aid in making clinical decisions for various pathologies like lung infection, diabetic retinopathy, etc.

Modified spider monkey optimization algorithm based feature selection and probabilistic neural network classifier in face recognition

AbstractThis paper proposes a novel and robust predictive method using modified spider monkey optimization (MSMO) and probabilistic neural network (PNN) for face recognition. The limitation of the traditional spider monkey optimization (SMO) approach to obtaining an optimal solution for classification problems is overcome by enhancing the performance of SMO by modifying the perturbation rate with a non‐linear function, thereby improving the convergence of SMO. The framework comprises image preprocessing, feature extraction using dual tree complex wavelet transform (DT‐CWT), feature selection using the modified spider monkey optimization algorithm (MSMO), and classification using PNN. The proposed method is tested on the Yale and AR Face datasets. Experimental outcomes reveal that the proposed framework attain an accuracy of 99.4% with appreciable sensitivity, specificity, and G‐mean. To examine the efficacy of MSMO, parametric studies are conducted, which showed that MSMO converges faster with high fitness when compared to similar evolutionary algorithms like Genetic Algorithm (GA), Grey Wolf Optimization Algorithm (GWO), Particle Swarm Algorithm (PSO), and Cuckoo Search (CS) in selecting the optimal feature set. The MSMO‐PNN method outperforms similar state‐of‐the‐art methods, which reveals that the method proposed is competitive. The proposed model is robust to Gaussian and salt–pepper noise, obtaining the highest accuracy of 97.89% for varied noise density and variance.

Improved swarm optimization of deep features for glaucoma classification using SEGSO and VGGNet

Efficient classification of glaucoma from fundus images remains crucial and a challenging task as the retinal anatomical structure is so complex in nature with varying contrast and boundaries. As a result, there is a chance that expert systems will misclassify the data. As a way to reduce the misclassification rate, a methodology wherein a deep learning approach integrated with an evolutionary algorithm is proposed. First, the relevant features are extracted using a pre-trained ImageNet model, the VGGNet. The features extorted are subsequently filtered using statistically enhanced Glow-worm Swarm Optimization (SEGSO). The algorithm identifies the most important and relevant features, while excluding those that are noisy or highly correlated. The efficiency of this fully automated system is evaluated on public and private retinal fundus image databases (totally 18,879 images) and the experiments demonstrated a high accuracy of 99.6% with less computational complexity, better sensitivity and specificity. SEGSO algorithm selected comparatively lesser features than the similar algorithms and when, employed with VGGNet outperform several other convolutional network models. The proposed method is robust against salt-pepper noise and achieved an accuracy of 97.2% when applied on degraded images. Feature extraction using Convolutional Neural Network and SEGSO feature optimization could prove to be a good combination to be used for other biomedical image classification processes.

Analisis Perbandingan Metode Harmonic Mean Filter dan Contraharmonic Mean Filter untuk mengurangi noise pada Citra Digital

The results of digital camera recordings to take digital images, there are often some disturbances that appear in the digital image called noise. The noise that is often found in digital images is Salt-Pepper Noise and Speckle Noise. An image that has noise usually occurs due to an error in image retrieval techniques. Noise that usually appears is spots on the image, it is necessary to reduce noise by using the right filter method so that the resulting image matches the original. One of the filter methods to reduce noise is the harmonic mean-filter and contra-harmonic mean filter methods. From the calculation results of the average Mean Square Error (MSE) and Peak Signal to Noise Ratio (PSNR), it can be concluded that the Harmonic Mean Filter method is better at reducing salt & pepper noise. While the Contra-Harmonic Mean Filter method is better at reducing speckle noise.

Reverse Image Search Using Deep Unsupervised Generative Learning and Deep Convolutional Neural Network

Reverse image search has been a vital and emerging research area of information retrieval. One of the primary research foci of information retrieval is to increase the space and computational efficiency by converting a large image database into an efficiently computed feature database. This paper proposes a novel deep learning-based methodology, which captures channel-wise, low-level details of each image. In the first phase, sparse auto-encoder (SAE), a deep generative model, is applied to RGB channels of each image for unsupervised representational learning. In the second phase, transfer learning is utilized by using VGG-16, a variant of deep convolutional neural network (CNN). The output of SAE combined with the original RGB channel is forwarded to VGG-16, thereby producing a more effective feature database by the ensemble/collaboration of two effective models. The proposed method provides an information rich feature space that is a reduced dimensionality representation of the image database. Experiments are performed on a hybrid dataset that is developed by combining three standard publicly available datasets. The proposed approach has a retrieval accuracy (precision) of 98.46%, without using the metadata of images, by using a cosine similarity measure between the query image and the image database. Additionally, to further validate the proposed methodology’s effectiveness, image quality has been degraded by adding 5% noise (Speckle, Gaussian, and Salt pepper noise types) in the hybrid dataset. Retrieval accuracy has generally been found to be 97% for different variants of noise

Curvature Best Basis: A Novel Criterion to Dynamically Select a Single Best Basis as the Extracted Feature for Periocular Recognition

Aiming at the problems in the best basis selection, this paper presents a novel criterion based on the statistical measurement of the curvature wavelet coefficient to dynamically select the single best basis of the quad-tree wavelet packet transformation. The selected single best basis works as an extracted feature for biometric periocular recognition system. The proposed method first extracts the mean curvature of wavelet coefficients inside the quad-tree wavelet packet transform. Second, the method finds the most distinctive features based on the largest standard deviation and dynamically selects the extracted curvature wavelet coefficients as the single best basis. Third, the selected single curvature best basis works as an extracted feature, and then it is combined with the histogram of oriented gradients method. Finally, the support vector machine is employed to perform classification. Two datasets of two-dimensional periocular digital images are tested against the proposed method. To show the extended ability, we analyze the curvature best basis method against wavelet functions and characteristics and test the proposed method against the plain face and masked face recognition. The proposed method achieves the highest performance results inside periocular recognition (97.53% accuracy for UBIPr-1 and 97.77% accuracy for EYB-P1), masked face recognition (98.11% accuracy), and plain face recognition (98.26% accuracy). The proposed method is robust against glasses occlusion, artificial geometry transformations, Gaussian and salt pepper noise. Comparison with other works in a similar recognition system shows that our proposed curvature best basis method yields the highest performance results.

Edge Defect Detection of Network Image by the Application of Modal Symmetry

In order to improve the accuracy of detection the image defect, a method to detect the edge defect based on modal symmetry algorithm was put forward. The improved PCNN was used to deal with the salt-pepper noise and Gaussian noise in image. On this basis, the semantic learning and annotation of image features were achieved. At first, the corresponding features were extracted from the original image. And then, the semantics were learned by combining the extracted features and the manually labeled library. Combined with the semantic annotation of image, the modal symmetry algorithm was adopted to linearly subtract the data collected by two centrosymmetric sampling points and thus to get the mean value. The asymmetric modal information of the whole image was obtained. Thus, the asymmetric modal could be extracted from the symmetrical modal. Due to the high amplitude of asymmetrical modal signal in defect location. Finally, the defect identification for various locations in image was completed by judging whether the amplitude of asymmetrical modal at the defect location had a sudden change. Following conclusions can be drawn from experimental results. The proposed method has excellent performance in image processing. Meanwhile, this method has high detection accuracy and practicability.

A Robust DWT-Based Compressed Domain Video Watermarking Technique

To achieve robustness and imperceptibility, an adaptive compressed domain blind video watermarking method based on Discrete Wavelet Transform (DWT) is proposed in this research. In this technique, multiple binary images derived from a single watermark image are first embedded in a video sequence. The spatial spread spectrum watermark is directly incorporated in the compressed bit streams by modifying the four sets of discrete wavelet coefficients. Comprehensive simulation experiments demonstrate that the developed approach is efficient and also robust against spatial attacks such as scaling and frame averaging, noise attacks such as Gaussian and salt pepper noise, and temporal attacks like frame dropping and shifting. Moreover, the proposed approach can also withstand against rotation attacks of arbitrary angle.

Adaptive Robust Blind Watermarking Scheme Improved by Entropy-Based SVM and Optimized Quantum Genetic Algorithm

With the intensive study of machine learning in digital watermarking, its ability to balance the robustness and transparency of watermarking technology has attracted researchers’ attention. Therefore, quantum genetic algorithm, which serves as an intelligent optimized scheme combined with biological genetic mechanism and quantum computing, is widely used in various fields. In this study, an adaptive robust blind watermarking algorithm by means of optimized quantum genetics (OQGA) and entropy classification-based SVM (support vector machine) is proposed. The host image was divided into two parts according to the odd and even rows of the host image. One part was transformed by DCT (discrete cosine transform), and then the embedding intensity and position were separately trained by entropy-based SVM and OQGA; the other part was by DWT (discrete wavelet transform), in which the key fusion was achieved by an ergodic matrix to embed the watermark. Simulation results indicate the proposed algorithm ensures the watermark scheme transparency as well as having better resistance to common attacks such as lossy JPEG compression, image darken, Gaussian low-pass filtering, contrast decreasing, salt-pepper noise, and geometric attacks such as rotation and cropping.

Chaotic Encryption and Privilege Based Visual Secret Sharing Model for Color Images

In the Privilege-based Visual Secret Sharing Model (PVSSM), each share has a unique privilege and a higher-privilege share contributes with more privilege to reveal the secret image. However, in PVSSM, when several shares with the higher priority are stacked, the secret image can be visibly displayed. This security problem is solved by applying a two-dimensional Logistic-Adjusted Sine Map (2D-LASM) to each share. This method is called Chaotic Encryption-based PVSSM. In this paper, we aim to present how Chaotic Encryption-based PVSSM is applied to color images. In order to assess the efficiency of this method, histogram analysis, data loss attack, salt-pepper noise attack, differential attack, chi-square analysis and correlation analysis tests were applied. The performance of this method has been evaluated according to NCPR, UACI, PSNR, SSIM and CQM. The proposed method achieved a good test values and showed better results compared to similar studies in literature.

An improved block-based matching algorithm of copy-move forgery detection

Copy-move forgery is a common way of image tampering. Matching algorithm is the key step in copy-move forgery detection. Usually, the classical block-based matching algorithm (CBMA) can’t find all matched sub-blocks. In this paper, we propose an improved block-based matching algorithm (IBMA) to solve the problem. Firstly, we put the sum of feature vectors in the first column to get a new matrix. Secondly, the matrix is sorted by first column. Finally, every row of the matrix will search the following rows until the difference in the first column is larger than the threshold value. Experiment results show that the improved block-based matching algorithm is better than the classical block-based matching algorithm when an image was distorted by Gaussian noise, salt-pepper noise, or JPEG compression. The reason is that improved block-based matching algorithm can look for all matched sub-blocks, which makes copy-move forgery detection methods more robust.

Aggregation of Discrete Cosine Transform Digital Image Watermarking with Advanced Encryption Standard Technique

Signal is a function of one or more independent variables which contain some information. Signals are of different category depending upon their characteristics. They are classified as noise signal, deterministic signal, random signal, periodic signal, non-periodic signal, physically realizable and non-realizable, energy and power signals, analog and digital signals, etc. The Information carry by these signals are in different formats like text, image, audio, video, etc. Due to advance internet technology, it is very easy to transfer data from source to destination within second. Thus, issue of security and copyright protection of data increases. There are different methods and techniques used for providing security and copyright protection used for data. In this research, aggregation of two techniques i.e. 2D-DCT based digital watermarking for copyright protection and AES technique using 256 bits key for security of digital image against different attacks like as jpeg compression, salt pepper noise, gaussian noise, median filter, cropping, etc.