Search results for: convolution backprojection

Authors: Sergey Ermolin, Olga Ermolin

Abstract:

A Deep Convolution Neural Network with Triplet Loss is used to identify duplicate images in real-estate advertisements in the presence of image artifacts such as watermarking, cropping, hue/brightness adjustment, and others. The effects of batch normalization, spatial dropout, and various convergence methodologies on the resulting detection accuracy are discussed. For comparative Return-on-Investment study (per industry request), end-2-end performance is benchmarked on both Nvidia Titan GPUs and Intel’s Xeon CPUs. A new real-estate dataset from San Francisco Bay Area is used for this work. Sufficient duplicate detection accuracy is achieved to supplement other database-grounded methods of duplicate removal. The implemented method is used in a Proof-of-Concept project in the real-estate industry.

Keywords: visual recognition, convolutional neural networks, triplet loss, spatial batch normalization with dropout, duplicate removal, advertisement technologies, performance benchmarking

Procedia PDF Downloads 340

Authors: Jia Li, Xing Wei, Yuchen Hong, Yang Lu

Abstract:

Detecting vehicle behavior has always been the focus of intelligent transportation, but with the explosive growth of the number of vehicles and the complexity of the road environment, the vehicle behavior videos captured by traditional surveillance have been unable to satisfy the study of vehicle behavior. The traditional method of manually labeling vehicle behavior is too time-consuming and labor-intensive, but the existing object detection and tracking algorithms have poor practicability and low behavioral location detection rate. This paper proposes a vehicle behavior detection algorithm based on the dual-stream convolution network and the multi-dimensional video dynamic detection network. In the videos, the straight-line behavior of the vehicle will default to the background behavior. The Changing lanes, turning and turning around are set as target behaviors. The purpose of this model is to automatically mark the target behavior of the vehicle from the untrimmed videos. First, the target behavior proposals in the long video are extracted through the dual-stream convolution network. The model uses a dual-stream convolutional network to generate a one-dimensional action score waveform, and then extract segments with scores above a given threshold M into preliminary vehicle behavior proposals. Second, the preliminary proposals are pruned and identified using the multi-dimensional video dynamic detection network. Referring to the hierarchical reinforcement learning, the multi-dimensional network includes a Timer module and a Spacer module, where the Timer module mines time information in the video stream and the Spacer module extracts spatial information in the video frame. The Timer and Spacer module are implemented by Long Short-Term Memory (LSTM) and start from an all-zero hidden state. The Timer module uses the Transformer mechanism to extract timing information from the video stream and extract features by linear mapping and other methods. Finally, the model fuses time information and spatial information and obtains the location and category of the behavior through the softmax layer. This paper uses recall and precision to measure the performance of the model. Extensive experiments show that based on the dataset of this paper, the proposed model has obvious advantages compared with the existing state-of-the-art behavior detection algorithms. When the Time Intersection over Union (TIoU) threshold is 0.5, the Average-Precision (MP) reaches 36.3% (the MP of baselines is 21.5%). In summary, this paper proposes a vehicle behavior detection model based on multi-dimensional dynamic detection network. This paper introduces spatial information and temporal information to extract vehicle behaviors in long videos. Experiments show that the proposed algorithm is advanced and accurate in-vehicle timing behavior detection. In the future, the focus will be on simultaneously detecting the timing behavior of multiple vehicles in complex traffic scenes (such as a busy street) while ensuring accuracy.

Keywords: vehicle behavior detection, convolutional neural network, long short-term memory, deep learning

Procedia PDF Downloads 132

Authors: Kavyadevi N., Kaviya G., Gowsalya P., Janani M., Mohanraj S.

Abstract:

Hevea brasiliensis, also known as the rubber tree, is one of the foremost assets of crops in the world. One of the most significant advantages of the Rubber Plant in terms of air oxygenation is its capacity to reduce the likelihood of an individual developing respiratory allergies like asthma. To construct such a system that can properly identify crop diseases and pests and then create a database of insecticides for each pest and disease, we must first give treatment for the illness that has been detected. We shall primarily examine three major leaf diseases since they are economically deficient in this article, which is Bird's eye spot, algal spot and powdery mildew. And the recommended work focuses on disease identification on rubber tree leaves. It will be accomplished by employing one of the superior algorithms. Input, Preprocessing, Image Segmentation, Extraction Feature, and Classification will be followed by the processing technique. We will use time-consuming procedures that they use to detect the sickness. As a consequence, the main ailments, underlying causes, and signs and symptoms of diseases that harm the rubber tree are covered in this study.

Keywords: image processing, python, convolution neural network (CNN), machine learning

Procedia PDF Downloads 77

Authors: Jinan Fiaidhi, Sabah Mohammed

Abstract:

Cross-domain learning in the medical field is a research challenge as many conditions, like in oncology imaging, use different imaging modalities. Moreover, in most of the medical learning applications, the sample training size is relatively small. Although few-shot learning (FSL) through the use of a Siamese neural network was able to be trained on a small sample with remarkable accuracy, FSL fails to be effective for use in multiple domains as their convolution weights are set for task-specific applications. In this paper, we are addressing this problem by enabling FSL to possess the ability to shift across domains by designing a two-layer FSL network that can learn individually from each domain and produce a shared features map with extra modulation to be used at the second layer that can recognize important targets from mix domains. Our initial experimentations based on mixed medical datasets like the Medical-MNIST reveal promising results. We aim to continue this research to perform full-scale analytics for testing our cross-domain FSL learning.

Keywords: Siamese neural network, few-shot learning, meta-learning, metric-based learning, thick data transformation and analytics

Procedia PDF Downloads 58

Authors: Muhammad Ali Sarwar, Muhammad Farooq, Nayab Hassan, Hammad Hassan

Abstract:

Pakistan is highly recognized for its agriculture and is well known for producing substantial amounts of wheat, cotton, and sugarcane. However, some factors contribute to a decline in crop quality and a reduction in overall output. One of the main factors contributing to this decline is the presence of weed and its late detection. This process of detection is manual and demands a detailed inspection to be done by the farmer itself. But by the time detection of weed, the farmer will be able to save its cost and can increase the overall production. The focus of this research is to identify and classify the four main types of weeds (Small-Flowered Cranesbill, Chick Weed, Prickly Acacia, and Black-Grass) that are prevalent in our region’s major crops. In this work, we implemented three different deep learning techniques: YOLO-v5, Inception-v3, and Deep CNN on the same Dataset, and have concluded that deep convolutions neural network performed better with an accuracy of 97.45% for such classification. In relative to the state of the art, our proposed approach yields 2% better results. We devised the architecture in an efficient way such that it can be used in real-time.

Keywords: deep convolution networks, Yolo, machine learning, agriculture

Procedia PDF Downloads 119

Authors: Kaustubh Chakradeo, Michael Delves, Sofya Titarenko

Abstract:

Malaria is a serious disease which affects hundreds of millions of people around the world, each year. If not treated in time, it can be fatal. Despite recent developments in malaria diagnostics, the microscopy method to detect malaria remains the most common. Unfortunately, the accuracy of microscopic diagnostics is dependent on the skill of the microscopist and limits the throughput of malaria diagnosis. With the development of Artificial Intelligence tools and Deep Learning techniques in particular, it is possible to lower the cost, while achieving an overall higher accuracy. In this paper, we present a VGG-based model and compare it with previously developed models for identifying infected cells. Our model surpasses most previously developed models in a range of the accuracy metrics. The model has an advantage of being constructed from a relatively small number of layers. This reduces the computer resources and computational time. Moreover, we test our model on two types of datasets and argue that the currently developed deep-learning-based methods cannot efficiently distinguish between infected and contaminated cells. A more precise study of suspicious regions is required.

Keywords: convolution neural network, deep learning, malaria, thin blood smears

Procedia PDF Downloads 131

Authors: Leah Ning

Abstract:

This paper discusses the development of a machine learning algorithm that accurately detects metastatic breast cancer (cancer has spread elsewhere from its origin part) in selected images that come from pathology scans of lymph node sections. Being able to develop an accurate artificial intelligence (AI) algorithm would help significantly in breast cancer diagnosis since manual examination of lymph node scans is both tedious and oftentimes highly subjective. The usage of AI in the diagnosis process provides a much more straightforward, reliable, and efficient method for medical professionals and would enable faster diagnosis and, therefore, more immediate treatment. The overall approach used was to train a convolution neural network (CNN) based on a set of pathology scan data and use the trained model to binarily classify if a new scan were benign or malignant, outputting a 0 or a 1, respectively. The final model’s prediction accuracy is very high, with 100% for the train set and over 70% for the test set. Being able to have such high accuracy using an AI model is monumental in regard to medical pathology and cancer detection. Having AI as a new tool capable of quick detection will significantly help medical professionals and patients suffering from cancer.

Keywords: breast cancer detection, AI, machine learning, algorithm

Procedia PDF Downloads 92

Authors: Dahi Ghareab Abdelsalam Ibrahim

Abstract:

Calibration of step heights with high accuracy is needed for many applications in the industry. In general, step height consists of three bands: pass band, transition band (roll-off), and stop band. Abdelsalam used a convolution of the transfer functions of both Chebyshev type 2 and elliptic filters with WFF of the Fresnel transform in the frequency domain for producing a steeper roll-off with the removal of ripples in the pass band- and stop-bands. In this paper, we used a new method based on the Hamming window: band-pass filter for calibration of step heights in terms of perfect adjustment of pass-band, roll-off, and stop-band. The method is applied to calibrate a nominal step height of 40 cm. The step height is measured first by asynchronous dual-wavelength phase-shift interferometry. The measured step height is then calibrated by the simulation of the Hamming window: band-pass filter. The spectrum of the simulated band-pass filter is simulated at N = 881 and f0 = 0.24. We can conclude that the proposed method can calibrate any step height by adjusting only two factors which are N and f0.

Keywords: optical metrology, step heights, hamming window, band-pass filter

Procedia PDF Downloads 84

Authors: Abe D. Desta

Abstract:

This study suggests a deep learning-based method for identifying and following moving objects in surveillance video. The proposed method uses a fast regional convolution neural network (F-RCNN) trained on a substantial dataset of vehicle images to first detect vehicles. A Kalman filter and a data association technique based on a Hungarian algorithm are then used to monitor the observed vehicles throughout time. However, in general, F-RCNN algorithms have been shown to be effective in achieving high detection accuracy and robustness in this research study. For example, in one study The study has shown that the vehicle detection and tracking, the system was able to achieve an accuracy of 97.4%. In this study, the F-RCNN algorithm was compared to other popular object detection algorithms and was found to outperform them in terms of both detection accuracy and speed. The presented system, which has application potential in actual surveillance systems, shows the usefulness of deep learning approaches in vehicle detection and tracking.

Keywords: artificial intelligence, computer vision, deep learning, fast-regional convolutional neural networks, feature extraction, vehicle tracking

Procedia PDF Downloads 129

Authors: Bin Ji, Kai Xu, Shunyu Yao, Jingjing Liu, Ye Pan

Abstract:

Human pose estimation is a popular research area in computer vision for its important application in human-machine interface. In recent years, 2D human pose estimation based on convolution neural network has got great progress and development. However, in more and more practical applications, people often need to deal with tasks based on video. It’s not far-fetched for us to consider how to combine the spatial and temporal information together to achieve a balance between computing cost and accuracy. To address this issue, this study proposes a new spatiotemporal model, namely Spatiotemporal Net (STNet) to combine both temporal and spatial information more rationally. As a result, the predicted keypoints heatmap is potentially more accurate and spatially more precise. Under the condition of ensuring the recognition accuracy, the algorithm deal with spatiotemporal series in a decoupled way, which greatly reduces the computation of the model, thus reducing the resource consumption. This study demonstrate the effectiveness of our network over the Penn Action Dataset, and the results indicate superior performance of our network over the existing methods.

Keywords: convolutional long short-term memory, deep learning, human pose estimation, spatiotemporal series

Procedia PDF Downloads 150

Authors: Muhammad Azam, Micheal Olaolu Arowolo, Fei He, Mihail Popescu, Dong Xu

Abstract:

The number of biological papers is growing quickly, which means that the number of biological pathway figures in those papers is also increasing quickly. Each pathway figure shows extensive biological information, like the names of genes and how the genes are related. However, manually annotating pathway figures takes a lot of time and work. Even though using advanced image understanding models could speed up the process of curation, these models still need to be made more accurate. To improve gene name recognition from pathway figures, we applied a Siamese network to map image segments to a library of pictures containing known genes in a similar way to person recognition from photos in many photo applications. We used a triple loss function and a triplet spatial pyramid pooling network by combining the triplet convolution neural network and the spatial pyramid pooling (TSPP-Net). We compared VGG19 and VGG16 as the Siamese network model. VGG16 achieved better performance with an accuracy of 93%, which is much higher than OCR results.

Keywords: biological pathway, image understanding, gene name recognition, object detection, Siamese network, VGG

Procedia PDF Downloads 292

Authors: Jalal Karam

Abstract:

The methods of Fourier, Laplace, and Wavelet Transforms provide transfer functions and relationships between the input and the output signals in linear time invariant systems. This paper shows the equivalence among these three methods and in each case presenting an application of the appropriate (Fourier, Laplace or Wavelet) to the convolution theorem. In addition, it is shown that the same holds for a direct integration method. The Biorthogonal wavelets Bior3.5 and Bior3.9 are examined and the zeros distribution of their polynomials associated filters are located. This paper also presents the significance of utilizing wavelets as effective tools in processing speech signals for common multimedia applications in general, and for recognition and compression in particular. Theoretically and practically, wavelets have proved to be effective and competitive. The practical use of the Continuous Wavelet Transform (CWT) in processing and analysis of speech is then presented along with explanations of how the human ear can be thought of as a natural wavelet transformer of speech. This generates a variety of approaches for applying the (CWT) to many paradigms analysing speech, sound and music. For perception, the flexibility of implementation of this transform allows the construction of numerous scales and we include two of them. Results for speech recognition and speech compression are then included.

Keywords: continuous wavelet transform, biorthogonal wavelets, speech perception, recognition and compression

Procedia PDF Downloads 417

Authors: M. V. Chidananda Murthy, M. Z. Kurian, H. S. Guruprasad

Abstract:

Super-resolution is the technique of intelligently upscaling images, avoiding artifacts or blurring, and deals with the recovery of a high-resolution image from one or more low-resolution images. Single-image super-resolution is a process of obtaining a high-resolution image from a set of low-resolution observations by signal processing. While super-resolution has been demonstrated to improve image quality in scaled down images in the image domain, its effects on the Fourier-based technique remains unknown. Super-resolution substantially improved the spatial resolution of the patient LGE images by sharpening the edges of the heart and the scar. This paper aims at investigating the effects of single image super-resolution on Fourier-based and image based methods of scale-up. In this paper, first, generate a training phase of the low-resolution image and high-resolution image to obtain dictionary. In the test phase, first, generate a patch and then difference of high-resolution image and interpolation image from the low-resolution image. Next simulation of the image is obtained by applying convolution method to the dictionary creation image and patch extracted the image. Finally, super-resolution image is obtained by combining the fused image and difference of high-resolution and interpolated image. Super-resolution reduces image errors and improves the image quality.

Keywords: image dictionary creation, image super-resolution, LGE images, patch extraction

Procedia PDF Downloads 375

Authors: Omer Cahana, Maya Herman, Ofer Levi

Abstract:

Magnetic resonance imaging (MRI) is a long medical scan that stems from a long acquisition time. This length is mainly due to the traditional sampling theorem, which defines a lower boundary for sampling. However, it is still possible to accelerate the scan by using a different approach, such as compress sensing (CS) or parallel imaging (PI). These two complementary methods can be combined to achieve a faster scan with high-fidelity imaging. In order to achieve that, two properties have to exist: i) the signal must be sparse under a known transform domain, ii) the sampling method must be incoherent. In addition, a nonlinear reconstruction algorithm needs to be applied to recover the signal. While the rapid advance in the deep learning (DL) field, which has demonstrated tremendous successes in various computer vision task’s, the field of MRI reconstruction is still in an early stage. In this paper, we present an extension of the state-of-the-art model in MRI reconstruction -VarNet. We utilize VarNet by using dilated convolution in different scales, which extends the receptive field to capture more contextual information. Moreover, we simplified the sensitivity map estimation (SME), for it holds many unnecessary layers for this task. Those improvements have shown significant decreases in computation costs as well as higher accuracy.

Keywords: MRI, deep learning, variational network, computer vision, compress sensing

Procedia PDF Downloads 163

Authors: Sajeeha Ansar, Asad Ali Safi, Sheikh Ziauddin, Ahmad R. Shahid, Faraz Ahsan

Abstract:

Gliomas are most challenging and aggressive type of tumors which appear in different sizes, locations, and scattered boundaries. CNN is most efficient deep learning approach with outstanding capability of solving image analysis problems. A fully automatic deep learning based 2D-CNN model for brain tumor segmentation is presented in this paper. We used small convolution filters (3 x 3) to make architecture deeper. We increased convolutional layers for efficient learning of complex features from large dataset. We achieved better results by pushing convolutional layers up to 16 layers for HGG model. We achieved reliable and accurate results through fine-tuning among dataset and hyper-parameters. Pre-processing of this model includes generation of brain pipeline, intensity normalization, bias correction and data augmentation. We used the BRATS-2015, and Dice Similarity Coefficient (DSC) is used as performance measure for the evaluation of the proposed method. Our method achieved DSC score of 0.81 for complete, 0.79 for core, 0.80 for enhanced tumor regions. However, these results are comparable with methods already implemented 2D CNN architecture.

Keywords: brain tumor segmentation, convolutional neural networks, deep learning, HGG

Procedia PDF Downloads 258

Authors: Ailing Yang, Penghan Song, Aihua Cai

Abstract:

The nonlinear maneuvering target tracking problem is mainly a state estimation problem when the target motion model is uncertain. Traditional solutions include Kalman filtering based on Bayesian filtering framework and extended Kalman filtering. However, these methods need prior knowledge such as kinematics model and state system distribution, and their performance is poor in state estimation of nonprior complex dynamic systems. Therefore, in view of the problems existing in traditional algorithms, a convolution LSTM target state estimation (SAConvLSTM-SE) algorithm based on Self-Attention memory (SAM) is proposed to learn the historical motion state of the target and the error distribution information measured at the current time. The measured track point data of airborne radar are processed into data sets. After supervised training, the data-driven deep neural network based on SAConvLSTM can directly obtain the target state at the next moment. Through experiments on two different maneuvering targets, we find that the network has stronger robustness and better tracking accuracy than the existing tracking methods.

Keywords: maneuvering target, state estimation, Kalman filter, LSTM, self-attention

Procedia PDF Downloads 180

Authors: Chaudhary Muhammad Aqdus Ilyas, Matthias Rehm, Kamal Nasrollahi, Thomas B. Moeslund

Abstract:

This paper presents the investigation of data collection procedures, associated with robots when placed with traumatic brain injured (TBI) patients for rehabilitation purposes through facial expression and mood analysis. Rehabilitation after TBI is very crucial due to nature of injury and variation in recovery time. It is advantageous to analyze these emotional signals in a contactless manner, due to the non-supportive behavior of patients, limited muscle movements and increase in negative emotional expressions. This work aims at the development of framework where robots can recognize TBI emotions through facial expressions to perform rehabilitation tasks by physical, cognitive or interactive activities. The result of these studies shows that with customized data collection strategies, proposed framework identify facial and emotional expressions more accurately that can be utilized in enhancing recovery treatment and social interaction in robotic context.

Keywords: computer vision, convolution neural network- long short term memory network (CNN-LSTM), facial expression and mood recognition, multimodal (RGB-thermal) analysis, rehabilitation, robots, traumatic brain injured patients

Procedia PDF Downloads 156

Authors: Cheng Fang, Lingwei Quan, Cunyue Lu

Abstract:

Human pose estimation and tracking are to accurately identify and locate the positions of human joints in the video. It is a computer vision task which is of great significance for human motion recognition, behavior understanding and scene analysis. There has been remarkable progress on human pose estimation in recent years. However, more researches are needed for human pose tracking especially for online tracking. In this paper, a framework, called PoseSRPN, is proposed for online single-person pose estimation and tracking. We use Siamese network attaching a pose estimation branch to incorporate Single-person Pose Tracking (SPT) and Visual Object Tracking (VOT) into one framework. The pose estimation branch has a simple network structure that replaces the complex upsampling and convolution network structure with deconvolution. By augmenting the loss of fully convolutional Siamese network with the pose estimation task, pose estimation and tracking can be trained in one stage. Once trained, PoseSRPN only relies on a single bounding box initialization and producing human joints location. The experimental results show that while maintaining the good accuracy of pose estimation on COCO and PoseTrack datasets, the proposed method achieves a speed of 59 frame/s, which is superior to other pose tracking frameworks.

Keywords: computer vision, pose estimation, pose tracking, Siamese network

Procedia PDF Downloads 154

Authors: Lindani Mbatha, Suvendi Rimer, Mpho Gololo

Abstract:

Image segmentation is the process of dividing an image into several sections, such as the object's background and the foreground. It is a critical technique in both image-processing tasks and computer vision. Most of the image segmentation algorithms have been developed for gray-scale images and little research and algorithms have been developed for the color images. Most image segmentation algorithms or techniques vary based on the input data and the application. Nearly all of the techniques are not suitable for noisy environments. Most of the work that has been done uses the Markov Random Field (MRF), which involves the computations and is said to be robust to noise. In the past recent years' image segmentation has been brought to tackle problems such as easy processing of an image, interpretation of the contents of an image, and easy analysing of an image. This article reviews and summarizes some of the image segmentation techniques and algorithms that have been developed in the past years. The techniques include neural networks (CNN), edge-based techniques, region growing, clustering, and thresholding techniques and so on. The advantages and disadvantages of medical ultrasound image segmentation techniques are also discussed. The article also addresses the applications and potential future developments that can be done around image segmentation. This review article concludes with the fact that no technique is perfectly suitable for the segmentation of all different types of images, but the use of hybrid techniques yields more accurate and efficient results.

Keywords: clustering-based, convolution-network, edge-based, region-growing

Procedia PDF Downloads 98

Authors: Amin Nezarat, Naeime Seifadini

Abstract:

Trip mode selection is a behavioral characteristic of passengers with immense importance for travel demand analysis, transportation planning, and traffic management. Identification of trip mode distribution will allow transportation authorities to adopt appropriate strategies to reduce travel time, traffic and air pollution. The majority of existing trip mode inference models operate based on human selected features and traditional machine learning algorithms. However, human selected features are sensitive to changes in traffic and environmental conditions and susceptible to personal biases, which can make them inefficient. One way to overcome these problems is to use neural networks capable of extracting high-level features from raw input. In this study, the convolutional neural network (CNN) architecture is used to predict the trip mode distribution based on raw GPS trajectory data. The key innovation of this paper is the design of the layout of the input layer of CNN as well as normalization operation, in a way that is not only compatible with the CNN architecture but can also represent the fundamental features of motion including speed, acceleration, jerk, and Bearing rate. The highest prediction accuracy achieved with the proposed configuration for the convolutional neural network with batch normalization is 85.26%.

Keywords: predicting, deep learning, neural network, urban trip

Procedia PDF Downloads 139

Authors: Hao Yuan, Yongjian Shen, Xiangjun He, Yuheng Li, Zhouzhou Zhang, Pengyu Zhang, Minkang Cai

Abstract:

Remote sensing data can provide support for decision-making in disaster assessment or disaster relief. The traditional processing methods of sensitive targets in remote sensing mapping are mainly based on manual retrieval and image editing tools, which are inefficient. Methods based on deep learning for sensitive target hiding are faster and more flexible. But these methods have disadvantages in training time and cost of calculation. This paper proposed a target hiding model Self Attention (SA) Deepfill, which used self-attention modules to replace part of gated convolution layers in image inpainting. By this operation, the calculation amount of the model becomes smaller, and the performance is improved. And this paper adds free-form masks to the model’s training to enhance the model’s universal. The experiment on an open remote sensing dataset proved the efficiency of our method. Moreover, through experimental comparison, the proposed method can train for a longer time without over-fitting. Finally, compared with the existing methods, the proposed model has lower computational weight and better performance.

Keywords: remote sensing mapping, image inpainting, self-attention mechanism, target hiding

Procedia PDF Downloads 139

Authors: Shalitha Jayasekara, Saluk Bawantha, Dinithi Anupama, Isuru Gunarathne, Pradeepa Bandara, Hansi De Silva

Abstract:

Dogs have been known as "man's best friend" for generations, providing friendship and loyalty to their human counterparts. However, due to people's busy lives, they are unaware of the ailments that can affect their pets. However, in recent years, mobile technologies have had a significant impact on our lives, and with technological improvements, a rule-based expert system allows the end-user to enable new types of healthcare systems. The advent of Android OS-based smartphones with more user-friendly interfaces and lower pricing opens new possibilities for continuous monitoring of pets' health conditions, such as healthy dogs, dangerous ingestions, and swallowed objects. The proposed ‘Best Bark’ Dog care and owner consultation system is a mobile application for dog owners. Four main components for dog owners were implemented after a questionnaire was distributed to the target group of audience and the findings were evaluated. The proposed applications are widely used to provide health and clinical support to dog owners, including suggesting exercise and diet plans and answering queries about their dogs. Additionally, after the owner uploads a photo of the dog, the application provides immediate feedback and a description of the dog's skin disease.

Keywords: Convolution Neural Networks, Artificial Neural Networks, Knowledgebase, Sentimental Analysis.

Procedia PDF Downloads 154

Authors: Xiang Yu, Chuntao Feng, Lu Yang, Meiyang Song, Wenhao Zhou

Abstract:

The existing two-dimensional micro-Doppler features extraction ignores the correlation information between the spatial and temporal dimension features. For the range-Doppler map, the time dimension is introduced, and a frequency modulation continuous wave (FMCW) radar human fall detection algorithm based on time-varying range-Doppler features is proposed. Firstly, the range-Doppler sequence maps are generated from the echo signals of the continuous motion of the human body collected by the radar. Then the three-dimensional data cube composed of multiple frames of range-Doppler maps is input into the three-dimensional Convolutional Neural Network (3D CNN). The spatial and temporal features of time-varying range-Doppler are extracted by the convolution layer and pool layer at the same time. Finally, the extracted spatial and temporal features are input into the fully connected layer for classification. The experimental results show that the proposed fall detection algorithm has a detection accuracy of 95.66%.

Keywords: FMCW radar, fall detection, 3D CNN, time-varying range-doppler features

Procedia PDF Downloads 123

Authors: L. Diogo, A. Legrand, J. Nguyen-Cao, J. Rogeau, S. Bornhofen

Abstract:

Lenia is a system of cellular automata with continuous states, space and time, which surprises not only with the emergence of interesting life-like structures but also with its beauty. This paper reports ongoing research on a GPU implementation of Lenia using CUDA C++ and OpenGL Interoperability. We demonstrate how CUDA as a low-level GPU programming paradigm allows optimizing performance and memory usage of the Lenia algorithm. A comparative analysis through experimental runs with existing implementations shows that the CUDA implementation outperforms the others by one order of magnitude or more. Cellular automata hold significant interest due to their ability to model complex phenomena in systems with simple rules and structures. They allow exploring emergent behavior such as self-organization and adaptation, and find applications in various fields, including computer science, physics, biology, and sociology. Unlike classic cellular automata which rely on discrete cells and values, Lenia generalizes the concept of cellular automata to continuous space, time and states, thus providing additional fluidity and richness in emerging phenomena. In the current literature, there are many implementations of Lenia utilizing various programming languages and visualization libraries. However, each implementation also presents certain drawbacks, which serve as motivation for further research and development. In particular, speed is a critical factor when studying Lenia, for several reasons. Rapid simulation allows researchers to observe the emergence of patterns and behaviors in more configurations, on bigger grids and over longer periods without annoying waiting times. Thereby, they enable the exploration and discovery of new species within the Lenia ecosystem more efficiently. Moreover, faster simulations are beneficial when we include additional time-consuming algorithms such as computer vision or machine learning to evolve and optimize specific Lenia configurations. We developed a Lenia implementation for GPU using the C++ and CUDA programming languages, and CUDA/OpenGL Interoperability for immediate rendering. The goal of our experiment is to benchmark this implementation compared to the existing ones in terms of speed, memory usage, configurability and scalability. In our comparison we focus on the most important Lenia implementations, selected for their prominence, accessibility and widespread use in the scientific community. The implementations include MATLAB, JavaScript, ShaderToy GLSL, Jupyter, Rust and R. The list is not exhaustive but provides a broad view of the principal current approaches and their respective strengths and weaknesses. Our comparison primarily considers computational performance and memory efficiency, as these factors are critical for large-scale simulations, but we also investigate the ease of use and configurability. The experimental runs conducted so far demonstrate that the CUDA C++ implementation outperforms the other implementations by one order of magnitude or more. The benefits of using the GPU become apparent especially with larger grids and convolution kernels. However, our research is still ongoing. We are currently exploring the impact of several software design choices and optimization techniques, such as convolution with Fast Fourier Transforms (FFT), various GPU memory management scenarios, and the trade-off between speed and accuracy using single versus double precision floating point arithmetic. The results will give valuable insights into the practice of parallel programming of the Lenia algorithm, and all conclusions will be thoroughly presented in the conference paper. The final version of our CUDA C++ implementation will be published on github and made freely accessible to the Alife community for further development.

Keywords: artificial life, cellular automaton, GPU optimization, Lenia, comparative analysis.

Procedia PDF Downloads 44

Authors: Anukriti Kumar, Tanmay Singh, Dinesh Kumar Vishwakarma

Abstract:

Traffic control has been one of the most common and irritating problems since the time automobiles have hit the roads. Problems like traffic congestion have led to a significant time burden around the world and one significant solution to these problems can be the proper implementation of the Intelligent Transport System (ITS). It involves the integration of various tools like smart sensors, artificial intelligence, position technologies and mobile data services to manage traffic flow, reduce congestion and enhance driver's ability to avoid accidents during adverse weather. Road and traffic signs’ recognition is an emerging field of research in ITS. Classification problem of traffic signs needs to be solved as it is a major step in our journey towards building semi-autonomous/autonomous driving systems. The purpose of this work focuses on implementing an approach to solve the problem of traffic sign classification by developing a Convolutional Neural Network (CNN) classifier using the GTSRB (German Traffic Sign Recognition Benchmark) dataset. Rather than using hand-crafted features, our model addresses the concern of exploding huge parameters and data method augmentations. Our model achieved an accuracy of around 97.6% which is comparable to various state-of-the-art architectures.

Keywords: multiclass classification, convolution neural network, OpenCV

Procedia PDF Downloads 177

Authors: Bowei Yuan, Shi Li, Liuyang Song, Huaqing Wang, Lingli Cui

Abstract:

Convolutional neural networks (CNN) have received extensive attention in the field of fault diagnosis. Many fault diagnosis methods use CNN for fault type identification. However, when the amount of raw data collected by sensors is massive, the neural network needs to perform a time-consuming classification task. In this paper, a mechanical fault diagnosis method based on vibration signal down-sampling and the improved one-dimensional convolutional neural network is proposed. Through the robust principal component analysis, the low-rank feature matrix of a large amount of raw data can be separated, and then down-sampling is realized to reduce the subsequent calculation amount. In the improved one-dimensional CNN, a smaller convolution kernel is used to reduce the number of parameters and computational complexity, and regularization is introduced before the fully connected layer to prevent overfitting. In addition, the multi-connected layers can better generalize classification results without cumbersome parameter adjustments. The effectiveness of the method is verified by monitoring the signal of the centrifugal pump test bench, and the average test accuracy is above 98%. When compared with the traditional deep belief network (DBN) and support vector machine (SVM) methods, this method has better performance.

Keywords: fault diagnosis, vibration signal down-sampling, 1D-CNN

Procedia PDF Downloads 133

Authors: Li Yu, Chenlu Jiao, Hongrun Xin, Yan Wang, Kaiyang Wang

Abstract:

In order to study the impact of various factors on the housing price, we propose to build different prediction models based on deep learning to determine the existing data of the real estate in order to more accurately predict the housing price or its changing trend in the future. Considering that the factors which affect the housing price vary widely, the proposed prediction models include two categories. The first one is based on multiple characteristic factors of the real estate. We built Convolution Neural Network (CNN) prediction model and Long Short-Term Memory (LSTM) neural network prediction model based on deep learning, and logical regression model was implemented to make a comparison between these three models. Another prediction model is time series model. Based on deep learning, we proposed an LSTM-1 model purely regard to time series, then implementing and comparing the LSTM model and the Auto-Regressive and Moving Average (ARMA) model. In this paper, comprehensive study of the second-hand housing price in Beijing has been conducted from three aspects: crawling and analyzing, housing price predicting, and the result comparing. Ultimately the best model program was produced, which is of great significance to evaluation and prediction of the housing price in the real estate industry.

Keywords: deep learning, convolutional neural network, LSTM, housing prediction

Procedia PDF Downloads 307

Authors: Jannes Quer, Amir Niknejad, Marcus Weber

Abstract:

Computational Drug Design is often based on Molecular Dynamics simulations of molecular systems. Molecular Dynamics can be used to simulate, e.g., the binding and unbinding event of a small drug-like molecule with regard to the active site of an enzyme or a receptor. However, the time-scale of the overall binding event is many orders of magnitude longer than the time-scale of simulation. Thus, there is a need to speed-up molecular simulations. In order to speed up simulations, the molecular dynamics trajectories have to be ”steared” out of local minimizers of the potential energy surface – the so-called metastabilities – of the molecular system. Increasing the kinetic energy (temperature) is one possibility to accelerate simulated processes. However, with temperature the entropy of the molecular system increases, too. But this kind ”stearing” is not directed enough to stear the molecule out of the minimum toward the saddle point. In this article, we give a new mathematical idea, how a potential energy surface can be changed in such a way, that entropy is kept under control while the trajectories are still steared out of the metastabilities. In order to compute the unsteared transition behaviour based on a steared simulation, we propose to use extrapolation methods. In the end we mathematically show, that our method accelerates the simulations along the direction, in which the curvature of the potential energy surface changes the most, i.e., from local minimizers towards saddle points.

Keywords: extrapolation, Eyring-Kramers, metastability, multilevel sampling

Procedia PDF Downloads 329

Authors: Mst Shapna Akter, Hossain Shahriar

Abstract:

Neural network approaches are machine learning methods used in many domains, such as healthcare and cyber security. Neural networks are mostly known for dealing with image datasets. While training with the images, several fundamental mathematical operations are carried out in the Neural Network. The operation includes a number of algebraic and mathematical functions, including derivative, convolution, and matrix inversion and transposition. Such operations require higher processing power than is typically needed for computer usage. Central Processing Unit (CPU) is not appropriate for a large image size of the dataset as it is built with serial processing. While Graphics Processing Unit (GPU) has parallel processing capabilities and, therefore, has higher speed. This paper uses advanced Neural Network techniques such as VGG16, Resnet50, Densenet, Inceptionv3, Xception, Mobilenet, XGBOOST-VGG16, and our proposed models to compare CPU and GPU resources. A system for classifying autism disease using face images of an autistic and non-autistic child was used to compare performance during testing. We used evaluation matrices such as Accuracy, F1 score, Precision, Recall, and Execution time. It has been observed that GPU runs faster than the CPU in all tests performed. Moreover, the performance of the Neural Network models in terms of accuracy increases on GPU compared to CPU.

Keywords: autism disease, neural network, CPU, GPU, transfer learning

Procedia PDF Downloads 121

Authors: Shital Suresh Borse, Vijayalaxmi Kadroli

Abstract:

E-commerce industries nowadays implement the latest AI, ML Techniques to improve their own performance and prediction accuracy. This helps to gain a huge profit from the online market. Ant Colony Optimization, Genetic algorithm, Particle Swarm Optimization, Neural Network & GWO help many e-commerce industries for up-gradation of their predictive performance. These algorithms are providing optimum results in various applications, such as stock price prediction, prediction of drug-target interaction & user ratings of similar products in e-commerce sites, etc. In this study, customer reviews will play an important role in prediction analysis. People showing much interest in buying a lot of services& products suggested by other customers. This ultimately increases net profit. In this work, a convolution neural network (CNN) is proposed which further is useful to optimize the prediction accuracy of an e-commerce website. This method shows that CNN is used to optimize hyperparameters of GWO algorithm using an appropriate coding scheme. Accurate model results are verified by comparing them to PSO results whose hyperparameters have been optimized by CNN in Amazon's customer review dataset. Here, experimental outcome proves that this proposed system using the GWO algorithm achieves superior execution in terms of accuracy, precision, recovery, etc. in prediction analysis compared to the existing systems.

Keywords: prediction analysis, e-commerce, machine learning, grey wolf optimization, particle swarm optimization, CNN

Procedia PDF Downloads 113