PCNN Model Research Articles

Federated deep learning algorithm for RPA network security detection

ABSTRACT The development of Internet technology has produced a huge impact on people’s lives. More data information is generated in life or work. Research proposes a combined detection model, and combines Fed Prox, convolutional neural network, and support vector machine to obtain Fed Prox Convolutional Neural Networks Support Vector Machine (Fed PCNN). The network intrusion detection model trains the local model with the divided local data, uses the model with the near-end item as the judgment basis, analyzes the difference between the model and the global model, and then trains the local model through local iteration. Finally, the local model is aggregated to obtain the best classification prediction effect of the model effect. The research proves that the Fed PCNN model has an excellent performance in recall and accuracy through experiments. In different data sets, the recall rate is more than 90%, the accuracy rate is more than 95%, and the false alarm rate is less than 2%. Therefore, the Fed PCNN model has a good detection effect in network intrusion detection.

PCNN Model Guided by Saliency Mechanism for Image Fusion in Transform Domain.

In heterogeneous image fusion problems, different imaging mechanisms have always existed between time-of-flight and visible light heterogeneous images which are collected by binocular acquisition systems in orchard environments. Determining how to enhance the fusion quality is key to the solution. A shortcoming of the pulse coupled neural network model is that parameters are limited by manual experience settings and cannot be terminated adaptively. The limitations are obvious during the ignition process, and include ignoring the impact of image changes and fluctuations on the results, pixel artifacts, area blurring, and the occurrence of unclear edges. Aiming at these problems, an image fusion method in a pulse coupled neural network transform domain guided by a saliency mechanism is proposed. A non-subsampled shearlet transform is used to decompose the accurately registered image; the time-of-flight low-frequency component, after multiple lighting segmentation using a pulse coupled neural network, is simplified to a first-order Markov situation. The significance function is defined as first-order Markov mutual information to measure the termination condition. A new momentum-driven multi-objective artificial bee colony algorithm is used to optimize the parameters of the link channel feedback term, link strength, and dynamic threshold attenuation factor. The low-frequency components of time-of-flight and color images, after multiple lighting segmentation using a pulse coupled neural network, are fused using the weighted average rule. The high-frequency components are fused using improved bilateral filters. The results show that the proposed algorithm has the best fusion effect on the time-of-flight confidence image and the corresponding visible light image collected in the natural scene, according to nine objective image evaluation indicators. It is suitable for the heterogeneous image fusion of complex orchard environments in natural landscapes.

Transmission Line Equipment Infrared Diagnosis Using an Improved Pulse-Coupled Neural Network

In order to detect the status of power equipment from infrared transmission line images under the spatial positioning relationship of the transmission line equipment, such as corridor, substation equipment, and facilities, this paper presents an improved PCNN model which merges an optimized parameter setting method. In this PCNN model, the original iteration mechanism is abandoned, and instead, the thresholding model is built by the maximum similarity thresholding rule. To ensure similarity during classifying neighboring neurons into cluster centers, a local clustering strategy is used for setting the linking coefficient, thus improving the efficiency of the method to detect the power equipment in infrared transmission line images. Finally, experimental results on transmission line infrared images show that the proposed method can provide the basis for the diagnosis of power equipment, preventing the casualties and property damage caused by the thermal damage of power equipment, and effectively improving the safety risk identification and operation control ability of power grid engineering.

Parallelistic Convolution Neural Network Approach for Brain Tumor Diagnosis.

Today, Magnetic Resonance Imaging (MRI) is a prominent technique used in medicine, produces a significant and varied range of tissue contrasts in each imaging modalities, and is frequently employed by medical professionals to identify brain malignancies. With brain tumor being a very deadly disease, early detection will help increase the likelihood that the patient will receive the appropriate medical care leading to either a full elimination of the tumor or the prolongation of the patient’s life. However, manually examining the enormous volume of magnetic resonance imaging (MRI) images and identifying a brain tumor or cancer is extremely time-consuming and requires the expertise of a trained medical expert or brain doctor to manually detect and diagnose brain cancer using multiple Magnetic Resonance images (MRI) with various modalities. Due to this underlying issue, there is a growing need for increased efforts to automate the detection and diagnosis process of brain tumor without human intervention. Another major concern most research articles do not consider is the low quality nature of MRI images which can be attributed to noise and artifacts. This article presents a Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm to precisely handle the problem of low quality MRI images by eliminating noisy elements and enhancing the visible trainable features of the image. The enhanced image is then fed to the proposed PCNN to learn the features and classify the tumor using sigmoid classifier. To properly train the model, a publicly available dataset is collected and utilized for this research. Additionally, different optimizers and different values of dropout and learning rates are used in the course of this study. The proposed PCNN with Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm achieved an accuracy of 98.7%, sensitivity of 99.7%, and specificity of 97.4%. In comparison with other state-of-the-art brain tumor methods and pre-trained deep transfer learning models, the proposed PCNN model obtained satisfactory performance.

Value of Artificial Neural Network Ultrasound in Improving Breast Cancer Diagnosis.

Ultrasound-guided needle biopsy based on artificial neural network, as a safe, effective, and simple preoperative pathological diagnosis technique, has been widely used in clinical practice. Ultrasound-guided needle biopsy based on artificial neural networks for suspicious breast lesions found in conventional ultrasound examinations is an effective method for preoperative diagnosis. The purpose of this article is to study the value of artificial neural network ultrasound in improving breast cancer diagnosis. This article summarizes the neuron model of PCNN by observing and studying its impulse synchronization phenomenon. Aiming at gray-scale images disturbed by mixed noise (impulse noise and the Gaussian noise), a comprehensive filtering algorithm based on the simplified PCNN model is proposed. In this paper, the benign and malignant breast masses were evaluated based on the two-dimensional and three-dimensional ultrasound imaging signs of the mass, and compared with the postoperative pathological results, a logistic regression model was established to analyze the shape, boundary, microcalcification, and posterior echo attenuation of the mass, values for keratinization or burrs, convergent signs, and blood flow classification in the differential diagnosis of benign and malignant. In this paper, a color ultrasound diagnostic device is used, Sonobi is used as a contrast medium, and the injection volume is 2.4 ml/dose. During the imaging process, the sound image performance of the lesion is dynamically observed, the original dynamic data are stored throughout the whole process, and the playback analysis is performed after the imaging is completed. Studies have shown that CDUS elastography (UE) combined with MRI can increase the sensitivity of breast cancer diagnosis, with a diagnostic accuracy rate of 92.4%.

Estimation of Ground PM2.5 Concentrations in Pakistan Using Convolutional Neural Network and Multi-Pollutant Satellite Images

During the last few decades, worsening air quality has been diagnosed in many cities around the world. The accurately prediction of air pollutants, particularly, particulate matter 2.5 (PM2.5) is extremely important for environmental management. A Convolutional Neural Network (CNN) P-CNN model is presented in this paper, which uses seven different pollutant satellite images, such as Aerosol index (AER AI), Methane (CH4), Carbon monoxide (CO), Formaldehyde (HCHO), Nitrogen dioxide (NO2), Ozone (O3) and Sulfur dioxide (SO2), as auxiliary variables to estimate daily average PM2.5 concentrations. This study estimates daily average of PM2.5 concentrations in various cities of Pakistan (Islamabad, Lahore, Peshawar and Karachi) by using satellite images. The dataset contains a total of 2562 images from May-2019 to April-2020. We compare and analyze AlexNet, VGG16, ResNet50 and P-CNN model on every dataset. The accuracy of machine learning models was checked with Mean Absolute Error (MAE), Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE). The results show that P-CNN is more accurate than other approaches in estimating PM2.5 concentrations from satellite images. This study presents robust model using satellite images, useful for estimating PM2.5 concentrations.

Application of CT Image Technology Based on Nearest Neighbor Propagation Clustering Segmentation Algorithm in Lung Cancer Radiotherapy

Objective. This paper uses the nearest neighbor propagation clustering segmentation algorithm to explore the impact of PET/CT image segmentation technology on lung cancer radiotherapy planning. Methods. In this paper, PET/CT scan was performed on 12 patients with nonmetastatic lung cancer. The self-written automatic segmentation program based on PCNN model is used to segment the PET target area, and then the tumor target area is manually sketched based on CT images and PET/CT images, and the intensity-modulated radiotherapy plan is formulated with the same parameters. Target volume and dose distribution were analyzed. Results. There was no statistical difference between the PET automatic segmentation target area and the PET manual contouring target area ( P < 0.05 ); the segmentation method was accurate and reliable; the difference between the CT manual contouring target area was statistically significant ( P 0.05 ). Conclusion. Based on the nearest neighbor propagation clustering segmentation algorithm, PET/CT image segmentation technology improves the accuracy of tumor target area delineation. The radiotherapy plan based on the segmentation target area can reduce the normal tissue exposure range and reduce the incidence of complications.

Design of Intelligent Education Service Robots for Big Data

With the development of robot technology, service robots have gradually come into people's lives, which also makes human-computer interaction more and more frequent. However, most of the current robot control algorithms have low accuracy and difficult operation, and people can't wait to find more effective algorithms. In order to design the most convenient and accurate intelligent education service robot, this paper uses SF algorithm and improved PCNN to establish a hybrid model, and proposes an improved saliency region extraction algorithm based on education service robot. The algorithm compares the standard database with the real environment, and the PR curve of the proposed algorithm is improved by about 5%. It is 10% higher than the SF algorithm in the AUC index, and the comprehensive F value is improved by 3.4% ~ 7.4%. This paper fully demonstrates that the saliency area generated by the proposed algorithm is closer to the true value, which can effectively suppress the high-brightness background area in the detection results of the SF algorithm. It also verifies that the PCNN model with the neuron propagation stimulation mechanism as the core more effective simulation of biological vision systems. Combining compressed sensing technology, this paper proposes a speech recognition scheme that is easy to implement in hardware. The algorithm performance of the robot is verified, and the optimal effect parameters are selected through comparative experiments. The method uses Chinese phonetic phrase (sentence) test to obtain good recognition results, and it can be used as an effective improvement scheme for the speech input of the proposed robot voice interaction system.

The Limited-FCMSPCNN algorithm for image defogging

In order to address the problem of defogging images of foggy scenes, this paper proposes a parameter-tunable image defogging algorithm based on the FC-MSPCNN model. Firstly, we simplify the traditional PCNN model and also propose the Limited-FCMSPCNN (LFC-MSPCNN) model. Secondly, we give the five key parameters in the model. There are five key parameters in the model: α, β, V, G and a modulation parameter q, and how they are set. Thirdly, we verified that the algorithm in this paper has good defogging effect on foggy sky images by comparing with the Dark Channel Prior algorithm, the Retinex algorithm and the FC-MSPCNN algorithm, showing that the LFC-MSPCNN method proposed in this paper has robustness.

Edge detection and location of seismic image based on PCNN

The discontinuous features in seismic data usually correspond to different geological edge information. Effective processing of these seismic data can help us make correct geological interpretations. In this paper, we apply coherent slices to detect and locate the boundary of fault polygons and propose a method for boundary detection and localization of seismic images based on PCNN. In image edge detection, the basic PCNN needs to adjust many parameters. This paper aims at this problem to improve the basic PCNN model, we simplify the feedback input and pulse input, only retain the external input stimulation and the connection domain external neuron stimulation, reduce the parameters, and simplify the calculation. Optimize parameters such as internal activity link coefficients, dynamic thresholds, and cycle times to improve interpretation efficiency and accuracy. Experiments show that the method used in this paper has good practicability and can effectively realize the fault polygons detection and positioning of coherent slices.

A method for enhancing speech and warning signals based on parallel convolutional neural networks in a noisy environment.

BACKGROUND: Digital hearing aids are based on technology that amplifies sound and removes noise according to the frequency of hearing loss in hearing loss patients. However, within the noise removed is a warning sound that alert the listener; the listener may be exposed to danger because the warning sound is not recognized.OBJECTIVE: In this paper, a deep learning model was used to improve these limits and propose a method to distinguish the warning sound in speech signals mixed with noise. In addition, the improved speech and warning sound were derived by removing noise present in the classification sound signals.METHODS: To classify the sound dataset, an adaptive convolution filter that changes according to two signals is proposed. The proposed convolution filter is applied to the PCNNs model to analyze the characteristics of the time and frequency domains of the dataset and classify the presence or absence of warning sound. In addition, the CEDN model was used to improve the intelligibility of the warning and the speech in the signal based on the warning sound classification from the proposed PCNNs model.RESULTS: Experimental results show that the PCNNs model using the proposed multiplicative filters is efficient for analyzing sound signals with complex frequencies. In addition, the CEDN model was used to improve the intelligibility of the warning and the speech in the signal based on the warning sound classification from the proposed PCNNs model.CONVLUSION: We confirmed that the PCNN model with the proposed filter showed the highest training rate, lowest error rate, and the most stable results. In addition, the CEDN model confirmed that speech and warning sounds were recognized, but it was confirmed that there was a limitation in clearly recognizing speech as the noise ratio increased.

An improved simplified PCNN model for salient region detection

As PCNN is modulated by using the pulse-coupled synaptic mechanisms, it has a great potential for image processing in a complex real-world environment, especially in images. A new simplified pulse coupled neural network (SPCNN) is proposed. This new model uses the pixel intensity with the actual physical meanings as the input parameters instead of the abstract network parameters in the original SPCNN. In order to achieve this goal, we try to derive the general formulae of dynamic threshold and internal activity of the SPCNN according to the dynamic properties of neurons and then deduce the relationship between the pixel intensity and the abstract parameters. Then, the relationship is transformed into an objective optimization problem to obtain the appropriate abstract parameters. Finally, extensive experiments are conducted on seven widely used datasets to demonstrate the effectiveness of the proposed method and shown improvement on the salient region detection.

Multi-Modality Image Fusion in Adaptive-Parameters SPCNN Based on Inherent Characteristics of Image

Due to the characteristics of the diversity and complement of multi-modality images, the fused method in SPCNN of adaptive parameters based on inherent characteristics of image is proposed in this paper, in order to extract the optimal feature information from the different images. Based on the analysis of traditional pulse coupled neural network, the simplified PCNN model based on SCM is designed. In order to highlight the different feature of different modality images, the adaptive parameters of SPCNN are set on the basis of the inherent characteristics of images, and the fused method in NSST is proposed. Firstly, the source image is decomposed by NSST with multiple scales and multiple directions. In the high-frequency coefficients, the SPCNN with adaptive parameters is adopted to calculate the firing time of each coefficient, the high-frequency coefficients are fused by comparing the firing times. In order to protect the details and edge information of the low-frequency coefficients, the combining of regional energy and energy of gradient is regarded as the active level, the low-frequency coefficients is fused by the weighted fusion rule. Finally, the fused image is reconstructed by inverse NSST transform. The experimental results show that the proposed method is superior to the other five classical methods, the fused results of the proposed method are consistence in the human visual perception system. The contrast and detail of fused image are perfect.

RETRACTED: Remote sensing image landmark segmentation algorithm based on improved GSA and PCNN combination

Application of Remote Sensing Technology in the Development of Urban Functions is presented. Remote sensing image processing for ecological protection and monitoring, this paper proposes a remote sensing image landmark segmentation algorithm based on the IGSA and PCNN. Because the GSA algorithm has the disadvantages of premature convergence and is easy to fall into the local optimal solution, the improved IGSA algorithm is used to extract the ratio of the image entropy and energy to IGSA, and the entropy change value is used as the IGSA algorithm. Based on the global search ability of IGSA, the optimal value of the key parameters affecting the segmentation effect in PCNN model is found. Finally, through experimental comparison, the proposed method has strong advantages in segmentation effect, real-time and robustness

A Review of True Orthophoto Rectification Algorithms

Orthophoto is one of the most important basic geographic information products. Because traditional orthophotos use incomplete surface information for digital differential rectification, there is mutual occlusion between images, which is difficult to meet the current large-scale urban mapping requirements. With the rapid development of photogrammetry technology and the increasing demand for high-precision graphics, the true orthophoto has become a major research hotspot at home and abroad. This paper expounds the development of the true orthophoto rectification, summarizes the relevant algorithms for the true orthophoto production, and points out the core problems and future development of the true orthophoto production. Through the summary and analysis of the image rectification algorithm, the image occlusion detection, shadow compensation and other issues are proposed. The morphological filtering algorithm is used to identify buildings in densely populated areas, and the PCNN model method enhances texture features, and combing the deep learning for shadow compensation, in order to optimize the actual image rectification algorithm for existing images.

Retinal blood vessel segmentation from diabetic retinopathy images using tandem PCNN model and deep learning based SVM

Diabetic Retinopathy (DR) occurs due to Type-II diabetes. It causes damages to the retinal blood vessels and reason for visual impairment. The predicted center is around the probability of variation in the estimation of retinal veins, and the crisp enrolls vessel development inside the retina. To witness the changes segmentation of retinal blood vessels has to be made. A framework to upgrade the quality of the segmentation results over morbid retinal images is proposed. This framework utilizes Contrast Limited Adaptive Histogram Equalization (CLAHE) for eliminating the background from the source image and enhances the foreground blood vessel pixels, Tandem Pulse Coupled Neural Network (TPCNN) model is endorsed for automatic feature vectors generation, and Deep Learning Based Support Vector Machine (DLBSVM) is proposed for classification and extraction of blood vessels. The DLBSVM parameters are fine-tuned via Firefly algorithm. The STARE, DRIVE, HRF, REVIEW, and DRIONS fundus image datasets are deliberated to assess the recommended techniques. The results render that the proposed technologies improve the segmentation with 80.61% Sensitivity, 99.54% Specificity, and 99.49% Accuracy.

Retinal Blood Vessel Segmentation from Depigmented Diabetic Retinopathy Images

Diabetic Retinopathy is a progressive disease that affects diabetic patients and changes the width and tortuosity of the retinal blood vessels. The preferred center of attention is to predict the new vessel growth and the dissimilarity in diameter of the retinal blood vessels. To examine the changes, primarily segmentation has to be made. A system has been proposed to enhance the quality of the segmentation result over pathological retinal images. The proposed system comprises preprocessing of Fundus images and extracts the blood vessels. The proposed system uses Contrast Limited Adaptive Histogram Equalization (CLAHE) for preprocessing and Tandem Pulse Coupled Neural Network (TPCNN) model to segment the retinal vasculature. To categorize the small blood vessels from pathological images, the algorithm depending on its parameters. In the former PCNN model, the parameters have to be set at every time for all images. The proposed TPCNN model assigns values for its multiple parameters through Particle Swarm Optimization (PSO); so that the decay speeds of the threshold would be regulated adaptively. This greatly enhances the flexibility of TPCNN in dealing with depigmented pathological images. The generated feature vectors of blood vessels are classified and extracted via Deep Learning Based Support Vector Machine (DLBSVM) technique. The proposed method is assessed over DRIVE, STARE, HRF, REVIEW, CHASE_DB1 and DRIONS databases by the performance parameters such as Sensitivity, Specificity, Accuracy, and Receiver Operating Characteristic (ROC) curve. The results render that these techniques improve the segmentation with an average value of 94.68% Sensitivity, 99.70% Specificity, 99.61% Accuracy and 98% ROC. The results evoke that the proposed methods are a suitable alternative for the supervised methods.

Dynamic threshold neural P systems

Pulse coupled neural networks (PCNN, for short) are models abstracting the synchronization behavior observed experimentally for the cortical neurons in the visual cortex of a cat’s brain, and the intersecting cortical model is a simplified version of the PCNN model. Membrane computing (MC) is a kind computation paradigm abstracted from the structure and functioning of biological cells that provide models working in cell-like mode, neural-like mode and tissue-like mode. Inspired from intersecting cortical model, this paper proposes a new kind of neural-like P systems, called dynamic threshold neural P systems (for short, DTNP systems). DTNP systems can be represented as a directed graph, where nodes are dynamic threshold neurons while arcs denote synaptic connections of these neurons. DTNP systems provide a kind of parallel computing models, they have two data units (feeding input unit and dynamic threshold unit) and the neuron firing mechanism is implemented by using a dynamic threshold mechanism. The Turing universality of DTNP systems as number accepting/generating devices is established. In addition, an universal DTNP system having 109 neurons for computing functions is constructed.

Remote sensing image fusion via RPCA and adaptive PCNN in NSST domain

In order to improve fused image quality of multi-spectral (MS) image and panchromatic (PAN) image, a new remote sensing image fusion algorithm based on robust principal component analysis (RPCA) and non-subsampled shearlet transform (NSST) is proposed. First, the first principle component PC1 of MS image is extracted via principal component analysis (PCA). Then, the component PC1 and PAN image are decomposed by NSST to get the low and high frequency subbands, respectively. For the low frequency subband, the sparse matrix of PAN image by RPCA decomposition is used to guide the fusion rule; for the high frequency subbands, the fusion rule employed is based on adaptive PCNN model. Finally, the fusion image is obtained by inverse NSST transform and inverse PCA transform. The experimental results illustrate that the proposed fusion algorithm outperforms other classical fusion algorithms (PCA, Curvelet, NSCT, NSST and NSCT-PCNN) in terms of visual quality and objective evaluation in whole, and achieve better fusion performance.

Medical Image Denoising using Unit Linking PCNN Model

Medical Image Denoising using Unit Linking PCNN Model