Image Processing Neural Network Research Articles

Analyzing ECG using pre-trained image processing neural networks

Analyzing ECG using pre-trained image processing neural networks

Automatic Outdoor Patrol Robot Based on Sensor Fusion and Face Recognition Methods

This study integrates path planning, fuzzy theory, neural networks, image processing, range sensors, webcam, global navigation satellite system (GNSS), and real-time kinematic (RTK) positioning system into an intelligent wheeled mobile robot (WMR) for outdoor patrolling. The robot system uses ultrasound sensors, laser sensors, and fuzzy controllers to detect obstacles and avoid them. The starting position and the goal position of the WMR in an outdoor environment are given by the GNSS RTK positioning system. Real-time position correction of the robot is performed through the differential global positioning system. The robot system applies the ant algorithm and the Dijkstra algorithm to find the shortest path for patrol tasks. The convolutional neural network image processing is utilized to identify intruders that are appearing in the patrol path. When the WMR detects an intruder by the face detection and recognition methods, the robot captures the photo of this person by the webcam and acquires the location information of this person by the RTK positioning system. Then the WMR sends the location and photo of the intruder to the control center by Wi-Fi and asks for help.

Back Propagation Neural Network-Based Ultrasound Image for Diagnosis of Cartilage Lesions in Knee Osteoarthritis.

Objective To explore the application value of ultrasound image based on back propagation (BP) neural network algorithm in knee osteoarthritis (KOA) and evaluate the application effect and value of ultrasound image technology based on the BP neural network in the diagnosis of knee osteoarthritis cartilage lesions, 98 patients who were admitted to our hospital were diagnosed with KOA and had undergone arthroscopic soft tissue examinations were randomly selected. According to whether image processing was performed, the ultrasound images of all patients were divided into two groups. The control group was image before processing, and the experimental group was image after processing optimization. The consistency of the inspection results of the ultrasound images before and after the processing with the arthroscopy results was compared. The results showed that the staging accuracy of the control group was 68.3% and that of the experimental group was 76.9%. The accuracy of staging cartilage degeneration of the experimental group was higher than that of the control group, and the difference was not remarkable (P > 0.05). The kappa coefficient of the experimental group was 0.61, and that of the control group was 0.40. The kappa coefficient of the experimental group was higher than that of the control group, and the difference was significant (P < 0.05). Conclusion The inspection effect of the ultrasound image processed by the BP neural network was superior to that of the conventional ultrasound image. It reflected the good adoption prospect of neural networks in image processing.

Watermarking Deep Neural Networks in Image Processing.

Publishing/sharing pretrained deep neural network (DNN) models is a common practice in the community of computer vision. The increasing popularity of pretrained models has made it a serious concern: how to protect the intellectual properties of model owners and avert illegal usages by malicious attackers. This article aims at developing a framework for watermarking DNNs, with a particular focus on low-level image processing tasks that map images to images. Using image denoising and superresolution as case studies, we develop a black-box watermarking method for pretrained models, which exploits the overparameterization of the DNNs in image processing. In addition, an auxiliary module for visualizing the watermark information is proposed for further verification. Extensive experiments show that the proposed watermarking framework has no noticeable impact on model performance and enjoys the robustness against the often-seen attacks.

Brain Tumor Classification Using Convolution Neural Network

Brain tumour is a very serious brain cancer. It is present or become due to the separation of the brain cells. In the recent field of this study, tells us that deep learning will help in health industry of medical diseases imaging in the Medical Diagnostic of all the diseases. CNN is mostly used in this Machine learning algorithm. Likewise, in this paper also, we bring out the Convolution Neural Network algorithm, image processing and data augmentation to say the brain images are cancerous and which are not cancerous. This project will require less computational power due to the transfer learning compared to the old CNN model. This has good accuracy results than the old pre-trained models.

Autonomous damage recognition in visual inspection of laminated composite structures using deep learning

This study proposes the exploitation of deep learning for quantitative assessment of visual detectability of different types of in-service damage in laminated composite structures such as aircraft and wind turbine blades. A comprehensive image-based data set is collected from the literature containing common microscale damage mechanisms (matrix cracking and fibre breakage) and macroscale damage mechanisms (impact and erosion). Then, automated classification of the damage type and severity was done by pre-trained version of AlexNet that is a stable convolutional neural network for image processing. Pre-trained ResNet-50 and 5 other user-defined convolutional neural networks were also used to evaluate the performance of AlexNet. The results demonstrated that employing AlexNet network, using the relatively small image dataset, provided the highest accuracy level (87%–96%) for identifying the damage severity and types in a reasonable computational time. The generated knowledge and the collected image data in this paper will facilitate further research and development in the field of autonomous visual inspection of composite structures with the potential to significantly reduce the costs, health & safety risks and downtime associated with integrity assessment.

Deep convolutional neural network image processing method providing improved signal-to-noise ratios in electron holography

Abstract An image identification method was developed with the aid of a deep convolutional neural network (CNN) and applied to the analysis of inorganic particles using electron holography. Despite significant variation in the shapes of α-Fe2O3 particles that were observed by transmission electron microscopy, this CNN-based method could be used to identify isolated, spindle-shaped particles that were distinct from other particles that had undergone pairing and/or agglomeration. The averaging of images of these isolated particles provided a significant improvement in the phase analysis precision of the electron holography observations. This method is expected to be helpful in the analysis of weak electromagnetic fields generated by nanoparticles showing only small phase shifts.

Artistic Style Meets Artificial Intelligence

Abstract Recent developments in neural network image processing motivate the question, how these technologies might better serve visual artists. Research goals to date have largely focused on either pastiche interpretations of what is framed as artistic “style” or seek to divulge heretofore unimaginable dimensions of algorithmic “latent space,” but have failed to address the process an artist might actually pursue, when engaged in the reflective act of developing an image from imagination and lived experience. The tools, in other words, are constituted in research demonstrations rather than as tools of creative expression. In this article, the authors explore the phenomenology of the creative environment afforded by artificially intelligent image transformation and generation, drawn from autoethnographic reviews of the authors’ individual approaches to artificial intelligence (AI) art. They offer a post-phenomenology of “neural media” such that visual artists may begin to work with AI technologies in ways that support naturalistic processes of thinking about and interacting with computationally mediated interactive creation.

Artistic Style Meets Artificial Intelligence

Recent developments in neural network image processing motivate the question, how these technologies might better serve visual artists. Research goals to date have largely focused on either pastiche interpretations of what is framed as artistic “style” or seek to divulge heretofore unimaginable dimensions of algorithmic “latent space,” but have failed to address the process an artist might actually pursue, when engaged in the reflective act of developing an image from imagination and lived experience. The tools, in other words, are constituted in research demonstrations rather than as tools of creative expression. In this article, the authors explore the phenomenology of the creative environment afforded by artificially intelligent image transformation and generation, drawn from autoethnographic reviews of the authors’ individual approaches to artificial intelligence (AI) art. They offer a post-phenomenology of “neural media” such that visual artists may begin to work with AI technologies in ways that support naturalistic processes of thinking about and interacting with computationally mediated interactive creation.

Intelligent Perception System of Robot Visual Servo for Complex Industrial Environment.

Robot control based on visual information perception is a hot topic in the industrial robot domain and makes robots capable of doing more things in a complex environment. However, complex visual background in an industrial environment brings great difficulties in recognizing the target image, especially when a target is small or far from the sensor. Therefore, target recognition is the first problem that should be addressed in a visual servo system. This paper considers common complex constraints in industrial environments and proposes a You Only Look Once Version 2 Region of Interest (YOLO-v2-ROI) neural network image processing algorithm based on machine learning. The proposed algorithm combines the advantages of YOLO (You Only Look Once) rapid detection with effective identification of ROI (Region of Interest) pooling structure, which can quickly locate and identify different objects in different fields of view. This method can also lead the robot vision system to recognize and classify a target object automatically, improve robot vision system efficiency, avoid blind movement, and reduce the calculation load. The proposed algorithm is verified by experiments. The experimental result shows that the learning algorithm constructed in this paper has real-time image-detection speed and demonstrates strong adaptability and recognition ability when processing images with complex backgrounds, such as different backgrounds, lighting, or perspectives. In addition, this algorithm can also effectively identify and locate visual targets, which improves the environmental adaptability of a visual servo system

An Automated Pipeline for Image Processing and Data Treatment to Track Activity Rhythms of Paragorgia arborea in Relation to Hydrographic Conditions.

Imaging technologies are being deployed on cabled observatory networks worldwide. They allow for the monitoring of the biological activity of deep-sea organisms on temporal scales that were never attained before. In this paper, we customized Convolutional Neural Network image processing to track behavioral activities in an iconic conservation deep-sea species—the bubblegum coral Paragorgia arborea—in response to ambient oceanographic conditions at the Lofoten-Vesterålen observatory. Images and concomitant oceanographic data were taken hourly from February to June 2018. We considered coral activity in terms of bloated, semi-bloated and non-bloated surfaces, as proxy for polyp filtering, retraction and transient activity, respectively. A test accuracy of 90.47% was obtained. Chronobiology-oriented statistics and advanced Artificial Neural Network (ANN) multivariate regression modeling proved that a daily coral filtering rhythm occurs within one major dusk phase, being independent from tides. Polyp activity, in particular extrusion, increased from March to June, and was able to cope with an increase in chlorophyll concentration, indicating the existence of seasonality. Our study shows that it is possible to establish a model for the development of automated pipelines that are able to extract biological information from times series of images. These are helpful to obtain multidisciplinary information from cabled observatory infrastructures.

Optic Disc Segmentation Using Attention-Based U-Net and the Improved Cross-Entropy Convolutional Neural Network.

Medical image segmentation is an important part of medical image analysis. With the rapid development of convolutional neural networks in image processing, deep learning methods have achieved great success in the field of medical image processing. Deep learning is also used in the field of auxiliary diagnosis of glaucoma, and the effective segmentation of the optic disc area plays an important assistant role in the diagnosis of doctors in the clinical diagnosis of glaucoma. Previously, many U-Net-based optic disc segmentation methods have been proposed. However, the channel dependence of different levels of features is ignored. The performance of fundus image segmentation in small areas is not satisfactory. In this paper, we propose a new aggregation channel attention network to make full use of the influence of context information on semantic segmentation. Different from the existing attention mechanism, we exploit channel dependencies and integrate information of different scales into the attention mechanism. At the same time, we improved the basic classification framework based on cross entropy, combined the dice coefficient and cross entropy, and balanced the contribution of dice coefficients and cross entropy loss to the segmentation task, which enhanced the performance of the network in small area segmentation. The network retains more image features, restores the significant features more accurately, and further improves the segmentation performance of medical images. We apply it to the fundus optic disc segmentation task. We demonstrate the segmentation performance of the model on the Messidor dataset and the RIM-ONE dataset, and evaluate the proposed architecture. Experimental results show that our network architecture improves the prediction performance of the base architectures under different datasets while maintaining the computational efficiency. The results render that the proposed technologies improve the segmentation with 0.0469 overlapping error on Messidor.

Reinforcement Learning using Convolutional Neural Network for Game Prediction

The paper presents a Deep learning model for playing computer games with elevated level information utilizing Reinforcement learning learning. The games are activity restricted (like snakes, catcher, air-bandit and so on.). The implementation is progressive in three parts. The first part deals with a simple neural network, the second one with Deep Q network and further to increase the accuracy and speed of the algorithm, the third part consists of a model consisting of convolution neural network for image processing and giving outputs from the fully connected layers so as to estimate the probability of an action being taken based on information extracted from inputs where we apply Q-learning to determine the best possible move. The results are further analysed and compared to provide an overview of the improvements in each methods.

Fast deep neural networks for image processing using posits and ARM scalable vector extension

With the advent of image processing and computer vision for automotive under real-time constraints, the need for fast and architecture-optimized arithmetic operations is crucial. Alternative and efficient representations for real numbers are starting to be explored, and among them, the recently introduced posit$$^{\mathrm{TM}}$$ number system is highly promising. Furthermore, with the implementation of the architecture-specific mathematical library thoroughly targeting single-instruction multiple-data (SIMD) engines, the acceleration provided to deep neural networks framework is increasing. In this paper, we present the implementation of some core image processing operations exploiting the posit arithmetic and the ARM scalable vector extension SIMD engine. Moreover, we present applications of real-time image processing to the autonomous driving scenario, presenting benchmarks on the tinyDNN deep neural network (DNN) framework.

Comparative Analysis among Discrete Fourier Transform, K-Means and Artificial Neural Networks Image Processing Techniques Oriented on Quality Control of Assembled Tires

The present paper discusses a comparative application of image processing techniques, i.e., Discrete Fourier Transform, K-Means clustering and Artificial Neural Network, for the detection of defects in the industrial context of assembled tires. The used Artificial Neural Network technique is based on Long Short-Term Memory and Fully Connected neural networks. The investigations focus on the monitoring and quality control of defects, which may appear on the external surface of tires after being assembled. Those defects are caused from tires which are not properly assembled to their respective metallic wheel rim, generating deformations and scrapes which are not desired. The proposed image processing techniques are applied on raw high-resolution images, which are acquired by in-line imaging and optical instruments. All the described techniques, i.e., Discrete Fourier Transform, K-Means clustering and Long Short-Term Memory, were able to determine defected and acceptable external tire surfaces. The proposed research is taken in the context of an industrial project which focuses on the development of automated quality control and monitoring methodologies, within the field of Industry 4.0 facilities. The image processing techniques are thus meant to be adopted into production processes, giving a strong support to the in-line quality control phase.

Colour recognition using colour histogram feature extraction and K-nearest neighbour classifier

K-nearest neighbours (KNN) is a widely used neural network and machine learning classification algorithm. Recently, it has been used in the neural network and digital image processing fields. In this study, the KNN classifier is used to distinguish 12 different colours. These colours are black, blue, brown, forest green, green, navy, orange, pink, red, violet, white and yellow. Using colour histogram feature extraction, which is one of the image processing techniques, the features that distinguish these colours are determined. These features increase the effectiveness of the KNN classifier. The training data consist of saved frames and the test data are obtained from the video camera in real-time. The video consists of consecutive frames. The frames are 100 × 70 in size. Each frame is tested with K = 3,5,7,9 and the obtained results are recorded. In general, the best results are obtained when used K = 5.&#x0D; &#x0D; Keywords: KNN algorithm, classifier, application, neural network, image processing, developed, colour, dataset, colour recognition.

Notice of Violation of IEEE Publication Principles: Ground-Based Cloud Image Recognition System Based on Multi-CNN and Feature Screening and Fusion

The recognition of ground-based cloud images has rich application prospects in many aspects such as weather prediction, astronomical site selection, and meteorological observation. Affected by factors such as rotation and illumination, the traditional feature extraction method is difficult to accurately describe the features of cloud images, resulting in low accuracy of ground-based cloud image recognition, and cannot meet the requirements of practical applications. With the popularity of convolutional neural networks in image processing, ground-based cloud image recognition algorithms based on convolutional neural network have become a research focus. However, the features of the ground-based cloud image are relatively shallow, and the cloud texture and other features are seriously lost in the convolution process, and it is difficult to achieve a good recognition effect. This paper proposes a ground-based cloud image recognition system based on multi-scale convolutional neural network (Multi-CNN) and multilayer perceptron neural networks (MLP). The multi-level and multi-scale convolution feature extraction is performed through convolution layers of Multi-CNN, and the local features with strong resolving power are selected through the feature screening algorithm based on DP clustering. Finally, the local features are encoded and fused for cloud image classification based on MLP. Filed test results show that our method was superior to other tested network models in terms of the recognition accuracy of 94.8% under 9 classification. In addition, ablation experiments show that the multi-scale feature extraction, screening and local feature coding in this paper have a significant effect on improving the algorithm’s ability to distinguish different cloud images.

Implementation and Design of 32 Bit Floating-Point ALU on a Hybrid FPGA-ARM Platform

FPGAs have capabilities such as low power consumption, multiple I/O pins, and parallel processing. Because of these capabilities, FPGAs are commonly used in numerous areas that require mathematical computing such as signal processing, artificial neural network design, image processing and filter applications. From the simplest to the most complex, all mathematical applications are based on multiplication, division, subtraction, addition. When calculating, it is often necessary to deal with numbers that are fractional, large or negative. In this study, the Arithmetic Logic Unit (ALU), which uses multiplication, division, addition, subtraction in the form of IEEE754 32-bit floating-point number used to represent fractional and large numbers is designed using FPGA part of the Xilinx Zynq-7000 integrated circuit. The programming language used is VHDL. Then, the ALU designed by the ARM processor part of the same integrated circuit was sent by the commands and controlled.

Multiple Reconstruction Compression Framework based on PNG Image

It is shown that neural networks (NNs) achieve excellent performances in image compression and reconstruction. However, there are still many shortcomings in the practical application, which eventually lead to the loss of neural network image processing ability. Based on this, this paper proposes a joint framework based on neural network and zoom compression. The framework first encodes the incoming PNG or JPEG image information, and then the image is converted into binary input decoder to reconstruct the intermediate state image, next we import the intermediate state image into the zooming compressor and re-pressurize it, and reconstruct the final image. From the experimental results, this method can better process the digital image and suppress the reverse expansion problem, and the compression effect can be improved by 4 to 10 times as much as that of using RNN alone, showing better ability in application. In this paper, the method is transmitted over a digital image, the effect is far better than the existing compression method alone, the Human visual system can not feel the change of the effect.

Super-Resolution Based on Residual Dense Network for Agricultural Image

With the rapid development of convolutional neural network in image processing, its image processing performance is becoming more and more outstanding. With the development of agricultural informatization, our demand for effectively accessing to agricultural information is also increasing. For example, greenhouse image monitoring equipment for crops is only in a low level. In this paper, we propose a method to apply image super resolution to the processing of greenhouse monitoring images, and use residual dense network to enhance the learning of global and local features in agricultural images. Therefore, more detailed information can be obtained from the low resolution images. Experiments on the data sets are made from the greenhouse agriculture images, and experimental results show that our method can effectively enhance the quality of agricultural monitoring images. It can better meet the needs of agricultural monitoring.