Optical Memory and Neural Networks最新文献_第10页

ASE-UNet: An Orange Fruit Segmentation Model in an Agricultural Environment Based on Deep Learning ASE-UNet：基于深度学习的农业环境中橙色水果分割模型

IF 1

Optical Memory and Neural Networks Pub Date : 2023-12-22 DOI: 10.3103/S1060992X23040045

Changgeng Yu, Dashi Lin, Chaowen He

{"title":"ASE-UNet: An Orange Fruit Segmentation Model in an Agricultural Environment Based on Deep Learning","authors":"Changgeng Yu, Dashi Lin, Chaowen He","doi":"10.3103/S1060992X23040045","DOIUrl":"10.3103/S1060992X23040045","url":null,"abstract":"<p>Fruit picking robot requires a powerful vision system that can accurately identify the fruit on the tree. Accurate segmentation of orange fruit in orchards is challenging because of the complex environments due to the overlapping of fruits and occlusions from foliage. In this work, we proposed an image segmentation model called ASE-UNet based on the U-Net architecture, which can achieve accurate segmentation of oranges in complex environments. Firstly, the backbone network structure is improved to reduce the down-sampling rate of orange fruit images, thereby retaining more spatial detail information. Secondly, we introduced the Shape Feature Extraction Module (SFEM), which at enhancing the ability of the model to distinguish between the fruits and backgrounds, such as branches and leaves, by extracting shape and outline information from the orange fruit target. Finally, an attention mechanism was utilized to suppress background channel feature interference in the skip connection and improve the fusion of high-layer and low-layer features. We evaluate the proposed model on the orange fruit images dataset collected in the agricultural environment. The results showed that ASE-UNet achieves IoU, Precision, Recall, and <i>F</i><sub>1</sub>-scores of 90.03, 96.10, 93.45, and 94.75%, respectively, which outperform other semantic segmentation methods, such as U-Net, PSPNet, and DeepLabv3+. The proposed method effectively solves the problem of low accuracy fruit segmentation models in the agricultural environment and provides technical support for fruit picking robots.</p>","PeriodicalId":721,"journal":{"name":"Optical Memory and Neural Networks","volume":"32 4","pages":"247 - 257"},"PeriodicalIF":1.0,"publicationDate":"2023-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138988825","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Application of the Luminescent Carbon Nanoparticles for Optical Diagnostics of Structure-Inhomogeneous Objects at the Micro- and Nanoscales 发光碳纳米粒子在微米和纳米尺度结构非均质物体光学诊断中的应用

IF 1

Optical Memory and Neural Networks Pub Date : 2023-12-22 DOI: 10.3103/S1060992X23040069

O. Angelsky, A. Bekshaev, C. Zenkova, D. Ivanskyi, P. Maksymyak, V. Kryvetsky, Zhebo Chen

引用次数: 0

Application of the Luminescent Carbon Nanoparticles for Optical Diagnostics of Structure-Inhomogeneous Objects at the Micro- and Nanoscales 发光碳纳米粒子在微米和纳米尺度结构非均质物体光学诊断中的应用

IF 0.9

Optical Memory and Neural Networks Pub Date : 2023-12-22 DOI: 10.3103/s1060992x23040069

O. Angelsky, A. Bekshaev, C. Zenkova, D. Ivanskyi, P. Maksymyak, V. Kryvetsky, Zhebo Chen

引用次数: 0

Data Augmentation and Fine Tuning of Convolutional Neural Network during Training for Person Re-Identification in Video Surveillance Systems 用于视频监控系统中人员再识别的卷积神经网络在训练过程中的数据增强和微调

IF 1

Optical Memory and Neural Networks Pub Date : 2023-12-22 DOI: 10.3103/S1060992X23040124

S. Ye, R. Bohush, H. Chen, S. Ihnatsyeva, S. V. Ablameyko

{"title":"Data Augmentation and Fine Tuning of Convolutional Neural Network during Training for Person Re-Identification in Video Surveillance Systems","authors":"S. Ye, R. Bohush, H. Chen, S. Ihnatsyeva, S. V. Ablameyko","doi":"10.3103/S1060992X23040124","DOIUrl":"10.3103/S1060992X23040124","url":null,"abstract":"<p>A new image set, augmentation method and fine in-learning adjustment of convolutional neural networks (CNN) are proposed to increase the accuracy of CNN-based person re-identification. Unlike other known sets, we have used many video frames from external and internal surveillance systems shot at all seasons of the year to make up our PolReID1077 set of person images. The PolReID1077-forming samples are subjected to the cyclic shift, chroma subsampling, and replacement of a fragment by a reduced copy of another sample to get a wider range of images. The learning set generating technique is used to train a CNN. The training is carried out in two stages. The first stage is pre-training using the augmented data. At the second stage the original images are used to carry out fine-tuning of CNN weight coefficients to reduce in-learning losses and increase re-identification efficiency. The approach doesn’t allow the CNN to remember learning sets and decreases the chances of overfitting. Different augmentation methods, data sets and learning techniques are used in the experiments.</p>","PeriodicalId":721,"journal":{"name":"Optical Memory and Neural Networks","volume":"32 4","pages":"233 - 246"},"PeriodicalIF":1.0,"publicationDate":"2023-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139015710","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Review on Pest Detection and Classification in Agricultural Environments Using Image-Based Deep Learning Models and Its Challenges 基于图像的深度学习模型在农业环境中的害虫检测和分类及其挑战综述

IF 1

Optical Memory and Neural Networks Pub Date : 2023-12-22 DOI: 10.3103/S1060992X23040112

P. Venkatasaichandrakanth, M. Iyapparaja

{"title":"Review on Pest Detection and Classification in Agricultural Environments Using Image-Based Deep Learning Models and Its Challenges","authors":"P. Venkatasaichandrakanth, M. Iyapparaja","doi":"10.3103/S1060992X23040112","DOIUrl":"10.3103/S1060992X23040112","url":null,"abstract":"<p>Agronomic pests cause agriculture to incur financial losses because they diminish production, which lowers revenue. Pest control, essential to lowering these losses, involves identifying and eliminating this risk. Since it enables management to take place, identification is the fundamental component of control. Utilizing the pest’s traits, visual identification is done. These characteristics differ between animals and are intrinsic. Since identification is so difficult, specialists in the field handle most of the work, which concentrates the information. Researchers have developed various techniques for predicting crop diseases using images of infected leaves. While progress has been made in identifying plant diseases using different models and methods, new advancements and discussions still offer room for improvement. Technology can significantly improve global crop production, and large datasets can be used to train models and approaches that uncover new and improved methods for detecting plant diseases and addressing low-yield issues. The effectiveness of machine learning and deep learning for identifying and categorizing pests has been confirmed by prior research. This paper thoroughly examines and critically evaluates the many strategies and methodologies used to classify and detect pests or insects using deep learning. The paper examines the benefits and drawbacks of various methodologies and considers potential problems with insect detection via image processing. The paper concludes by providing an analysis and outlook on the future direction of pest detection and classification using deep learning on plants like peanuts.</p>","PeriodicalId":721,"journal":{"name":"Optical Memory and Neural Networks","volume":"32 4","pages":"295 - 309"},"PeriodicalIF":1.0,"publicationDate":"2023-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139013597","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Plant Foliage Disease Diagnosis Using Light-Weight Efficient Sequential CNN Model 利用轻量高效序列 CNN 模型诊断植物叶面病害

IF 1

Optical Memory and Neural Networks Pub Date : 2023-12-22 DOI: 10.3103/S1060992X23040100

Raj Kumar, Anuradha Chug, Amit Prakash Singh

{"title":"Plant Foliage Disease Diagnosis Using Light-Weight Efficient Sequential CNN Model","authors":"Raj Kumar, Anuradha Chug, Amit Prakash Singh","doi":"10.3103/S1060992X23040100","DOIUrl":"10.3103/S1060992X23040100","url":null,"abstract":"<p>The Precise and prompt identification of plant pathogens is essential to keep agricultural losses as low as possible. In recent time, deep convolution neural networks have seen an exponential growth in their use in phytopathology due to its capacity for rapid and precise disease identification. However, deep convolutional neural network needs a lot of processing power because of its intricate structure consisting of a large stack of layers and millions of trainable parameters which makes them inedquate for light computing devices. In this article, authors have introduced a novel light-weight sequential CNN architecture for the diagnosis of leaf diseases. The suggested CNN approach contains fewer layers and around 70% less attributes than pre-trained CNN-based approaches. For the experiments and performance evaluation, authors have chosen a benchmark public dataset consisting of 7012 images of tomato and potato leaves affected with early and late blight diseases. The performance of the proposed architecture is compared against three recent priorly trained CNN architectures such as ResNet-50, VGG-16 and MobileNet-V2. The average accuracy percentage reported by the proposed architecture is 98.02 and the time consumed in training is also much better than the existing priorly trained CNN architectures. The experimental findings clearly demonstrate that the suggested approach outperforms the recent existing trained CNN approaches and has a very less number of layers and parameters which significantly reduces the amount of computing resources and time to train the model which could be a better choice for mobile-based real-time plant disease diagnosis applications.</p>","PeriodicalId":721,"journal":{"name":"Optical Memory and Neural Networks","volume":"32 4","pages":"331 - 345"},"PeriodicalIF":1.0,"publicationDate":"2023-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139023049","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Data Augmentation and Fine Tuning of Convolutional Neural Network during Training for Person Re-Identification in Video Surveillance Systems 用于视频监控系统中人员再识别的卷积神经网络在训练过程中的数据增强和微调

IF 0.9

Optical Memory and Neural Networks Pub Date : 2023-12-22 DOI: 10.3103/s1060992x23040124

S. Ye, R. Bohush, H. Chen, S. Ihnatsyeva, S. V. Ablameyko

{"title":"Data Augmentation and Fine Tuning of Convolutional Neural Network during Training for Person Re-Identification in Video Surveillance Systems","authors":"S. Ye, R. Bohush, H. Chen, S. Ihnatsyeva, S. V. Ablameyko","doi":"10.3103/s1060992x23040124","DOIUrl":"https://doi.org/10.3103/s1060992x23040124","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>A new image set, augmentation method and fine in-learning adjustment of convolutional neural networks (CNN) are proposed to increase the accuracy of CNN-based person re-identification. Unlike other known sets, we have used many video frames from external and internal surveillance systems shot at all seasons of the year to make up our PolReID1077 set of person images. The PolReID1077-forming samples are subjected to the cyclic shift, chroma subsampling, and replacement of a fragment by a reduced copy of another sample to get a wider range of images. The learning set generating technique is used to train a CNN. The training is carried out in two stages. The first stage is pre-training using the augmented data. At the second stage the original images are used to carry out fine-tuning of CNN weight coefficients to reduce in-learning losses and increase re-identification efficiency. The approach doesn’t allow the CNN to remember learning sets and decreases the chances of overfitting. Different augmentation methods, data sets and learning techniques are used in the experiments.</p>","PeriodicalId":721,"journal":{"name":"Optical Memory and Neural Networks","volume":"120 1","pages":""},"PeriodicalIF":0.9,"publicationDate":"2023-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139029213","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Investigating the Efficiency of Using U-Net, Erf-Net and DeepLabV3 Architectures in Inverse Lithography-based 90-nm Photomask Generation 研究在基于反向光刻技术的 90 纳米光掩膜生成中使用 U-Net、Erf-Net 和 DeepLabV3 架构的效率

IF 1

Optical Memory and Neural Networks Pub Date : 2023-12-22 DOI: 10.3103/S1060992X23040094

I. M. Karandashev, G. S. Teplov, A. A. Karmanov, V. V. Keremet, A. V. Kuzovkov

引用次数: 0

Plant Foliage Disease Diagnosis Using Light-Weight Efficient Sequential CNN Model 利用轻量高效序列 CNN 模型诊断植物叶面病害

IF 0.9

Optical Memory and Neural Networks Pub Date : 2023-12-22 DOI: 10.3103/s1060992x23040100

Raj Kumar, Anuradha Chug, Amit Prakash Singh

{"title":"Plant Foliage Disease Diagnosis Using Light-Weight Efficient Sequential CNN Model","authors":"Raj Kumar, Anuradha Chug, Amit Prakash Singh","doi":"10.3103/s1060992x23040100","DOIUrl":"https://doi.org/10.3103/s1060992x23040100","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>The Precise and prompt identification of plant pathogens is essential to keep agricultural losses as low as possible. In recent time, deep convolution neural networks have seen an exponential growth in their use in phytopathology due to its capacity for rapid and precise disease identification. However, deep convolutional neural network needs a lot of processing power because of its intricate structure consisting of a large stack of layers and millions of trainable parameters which makes them inedquate for light computing devices. In this article, authors have introduced a novel light-weight sequential CNN architecture for the diagnosis of leaf diseases. The suggested CNN approach contains fewer layers and around 70% less attributes than pre-trained CNN-based approaches. For the experiments and performance evaluation, authors have chosen a benchmark public dataset consisting of 7012 images of tomato and potato leaves affected with early and late blight diseases. The performance of the proposed architecture is compared against three recent priorly trained CNN architectures such as ResNet-50, VGG-16 and MobileNet-V2. The average accuracy percentage reported by the proposed architecture is 98.02 and the time consumed in training is also much better than the existing priorly trained CNN architectures. The experimental findings clearly demonstrate that the suggested approach outperforms the recent existing trained CNN approaches and has a very less number of layers and parameters which significantly reduces the amount of computing resources and time to train the model which could be a better choice for mobile-based real-time plant disease diagnosis applications.</p>","PeriodicalId":721,"journal":{"name":"Optical Memory and Neural Networks","volume":"22 1","pages":""},"PeriodicalIF":0.9,"publicationDate":"2023-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139029382","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Development of Prediction Models for Vulnerable Road User Accident Severity 开发易受伤害道路使用者事故严重程度预测模型

IF 1

Optical Memory and Neural Networks Pub Date : 2023-12-22 DOI: 10.3103/S1060992X23040082

Saurabh Jaglan, Sunita Kumari, Praveen Aggarwal

{"title":"Development of Prediction Models for Vulnerable Road User Accident Severity","authors":"Saurabh Jaglan, Sunita Kumari, Praveen Aggarwal","doi":"10.3103/S1060992X23040082","DOIUrl":"10.3103/S1060992X23040082","url":null,"abstract":"<p>Road traffic accidents are considered a significant problem which ruins the life of many people and also causes major economic losses. So, this issue is considered a hot research topic, and many researchers all over the world are focusing on developing a solution to this most challenging problem. Traditionally the accident spots are detected by means of transportation experts, and following that, some of the statistical models such as linear and nonlinear regression were used for accident severity prediction. However, these traditional approaches do not have the capability to analyze the relationship between the influential factor and accident severity. To address this issue, an Artificial Neural Network (ANN) classifier based vulnerable accident prediction model is proposed in this current research. Initially, the past accident data over the past period of years is collected from a specified area. The acquired data consists of a variable factor related to road infrastructure, weather condition, area of the accident, type of injury and driving characteristics. Then, to standardize the raw input data, min-max normalization is used as a pre-processing technique. The pre-processed is sent for the feature selection process in which essential features are selected by correlating the variable factor with accident severity prediction. Following that, the dimension of the features is reduced using Latent Sematic Index (LSI). Finally, the reduced features are fetched into the ANN classifier for predicting the severity of accidents such as low, medium and high. Simulation analysis of the proposed accident prediction model is carried out by evaluating some of the performance metrics for three datasets. Accuracy, error, specificity, recall and precision attained for the proposed model using dataset 1 is 96.3, 0.03, 98 and 98%. Through this proposed vulnerable accident prediction model, the severity of accidents can be analyzed effectively, and road safety levels can be improved.</p>","PeriodicalId":721,"journal":{"name":"Optical Memory and Neural Networks","volume":"32 4","pages":"346 - 363"},"PeriodicalIF":1.0,"publicationDate":"2023-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139013537","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0