http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
A Study on Weeds Retrieval based on Deep Neural Network Classification Model
Vo Hoang Trong,Gwang-Hyun Yu,Dang Thanh Vu,Ju-Hwan Lee,Nguyen Huy Toan,Jin-Young Kim 한국정보기술학회 2020 한국정보기술학회논문지 Vol.18 No.8
In this paper, we study the ability of content-based image retrieval by extracting descriptors from a deep neural network (DNN) trained for classification purposes. We fine-tuned the VGG model for the weeds classification task. Then, the feature vector, also a descriptor of the image, is obtained from a global average pooling (GAP) and two fully connected (FC) layers of the VGG model. We apply the principal component analysis (PCA) and develop an autoencoder network to reduce the dimension of descriptors to 32, 64, 128, and 256 dimensions. We experiment weeds species retrieval problem on the Chonnam National University (CNU) weeds dataset. The experiment shows that collecting features from DNN trained for weeds classification task can perform well on image retrieval. Without applying dimensionality reduction techniques, we get 0.97693 on the mean average precision (mAP) value. Using autoencoder to reduced dimensional descriptors, we achieve 0.97719 mAP with the descriptor dimension is 256.
Vo Hoang Trong(보 호앙 트롱),Ji-Hoon Bae(배지훈),Gwang-Hyun Yu(유광현),Jin-Young Kim(김진영) 한국디지털콘텐츠학회 2021 한국디지털콘텐츠학회논문지 Vol.22 No.10
Initializing the weights plays an essential role in a convolutional neural network model. This paper investigates how Glorot and Hes initialization methods behave in Mobilenet and Resnet models on the weeds classification problem. Experiments show that pointwise and depthwise convolution in Mobilenet reduces the variance of feature maps from earlier layers. Using the He’s method, shortcut connection in Resnet saturate values in logistic classify layer. The accuracy of Mobilenet and Resnet, using Glorots method, are 0.9568 and 0.9711, respectively. While using Hes method, we obtain 0.9471 using Mobilenet and 0.9645 using Resnet. Also, both models converge faster and better generalization using Glorots method than using Hes method.
Vo Hoang Trong,Yu Gwang-hyun,Dang Thanh Vu,Lee Ju-hwan,Nguyen Huy Toan,Kim Jin-young 한국스마트미디어학회 2020 스마트미디어저널 Vol.9 No.4
In the image object classification problem, low-resolution images may have a negative impact on the classification result, especially when the classification method, such as a convolutional neural network (CNN) model, is trained on a high-resolution (HR) image dataset. In this paper, we analyze the behavior of applying a classical super-resolution (SR) method such as bicubic interpolation, and a deep CNN model such as SRCNN to enhance low-resolution (LR) weeds images used for classification. Using an HR dataset, we first train a CNN model for weeds image classification with a default input size of 128×128. Then, given an LR weeds image, we rescale to default input size by applying the bicubic interpolation or the SRCNN model. We analyze these two approaches on the Chonnam National University (CNU) weeds dataset and find that SRCNN is suitable for the image size is smaller than 80×80, while bicubic interpolation is convenient for a larger image.
Analyze weeds classification with visual explanation based on Convolutional Neural Networks
Vo, Hoang-Trong,Yu, Gwang-Hyun,Nguyen, Huy-Toan,Lee, Ju-Hwan,Dang, Thanh-Vu,Kim, Jin-Young THE KOREAN INSTITUTE OF SMART MEDIA 2019 스마트미디어저널 Vol.8 No.3
To understand how a Convolutional Neural Network (CNN) model captures the features of a pattern to determine which class it belongs to, in this paper, we use Gradient-weighted Class Activation Mapping (Grad-CAM) to visualize and analyze how well a CNN model behave on the CNU weeds dataset. We apply this technique to Resnet model and figure out which features this model captures to determine a specific class, what makes the model get a correct/wrong classification, and how those wrong label images can cause a negative effect to a CNN model during the training process. In the experiment, Grad-CAM highlights the important regions of weeds, depending on the patterns learned by Resnet, such as the lobe and limb on 미국가막사리, or the entire leaf surface on 단풍잎돼지풀. Besides, Grad-CAM points out a CNN model can localize the object even though it is trained only for the classification problem.
다양한 합성곱 신경망 접근법을 이용한 잡초 이미지 분류
보 호앙 트롱(Vo Hoang Trong),유광현(Gwang-Hyun Yu),나지르 샤히드(Nazeer Shahid),황성민(Seong-Min Hwang),김진영(Jin-Young Kim) 대한전자공학회 2018 대한전자공학회 학술대회 Vol.2018 No.11
In this paper, we present a multimodal approach for weeds classification. We apply the transfer learning to classify on Convolutional Neural Networks(CNN) VGG16, Inception-Resnet, and Mobilenet separately. Then, we combine probabilities returned from each model, and start voting by scoring classes. We choose a class that has the highest score to conclude the final classification. We experiment on own weeds dataset and achieve 95.927% accuracy after voting on fusion classification.
유광현,Vo Hoang Trong,김진영 한국통신학회 2021 정보와 통신 Vol.38 No.8
농산물의 생육 및 생산에 영향을 미치는 요소는 매우 다양하지만, 중요한 요인 중 하나는 잡초방제이다. 정밀한 잡초 방제를 위해서는 잡초의 식별 그리고 식별 잡초에 따른 정확한 방제법으로 제거해야 한다. 잡초는 일반적으로 꽃, 화서, 열매, 잎, 줄기, 그리고 뿌리의 형태적 특성에 따라 구분되고, 이러한 식별법에 따라 영상기반의 잡초 식별에 관한 연구가 이루어지고 있다. 본 고에서는 영상 식별 분야에서 가장 우수한 성능을 보이는 딥러닝 기반의 영상인식 연구 동향을 살펴보고, 딥러닝 기반 잡초 식별을 위한 국내·외의 데이터베이스 및 연구개발의 예들을 고찰한다.
딥러닝 기반 자동 사과 선별기 구축을 위한 사과 품질 분류 모델
이주환(Lee Ju Hwan),Vo Hoang Trong,김진영(Kim Jin Young) 한국통신학회 2022 한국통신학회 학술대회논문집 Vol.2022 No.2
인공지능 기반의 농업자동화 연구는 노동력 절감 및 생산성 향성을 위한 주요한 과제이다. 현재 과수원의 자동화 시스템은 기계식 선별기와 함께 인력에 의한 선별 작업이 필수적으로 요구되며 이는 많은 시간과 노동이 요구된다. 본 논문은 딥러닝 기반의 자동화 선별기 시스템 구축을 위한 사과 이미지를 임베디드 환경에서 빠르게 품질 분류할 수 있는 모델을 제안한다. 사과의 품질은 양품과 불량품의 기준으로 분류하며 모델 신뢰성 검증을 위한 CAM(Class Activation Mapping)을 통해 분류 결과를 시각화하여 모델을 검증한다. 제안된 모델은 임베디드 환경에서 93% 이상의 정확도와 0.2초의 속도로 사과의 품질 분류가 가능하며 향후 저비용의 높은 정확도와 속도를 갖춘 딥러닝 기반 자동 선별기 구축이 가능할 것으로 예상한다.
Removing Out - Of - Distribution Samples on Classification Task
Thanh-Vu Dang,Hoang-Trong Vo,유광현(Gwang-Hyun Yu),이주환(Ju-Hwan Lee),Huy-Toan Nguyen,김진영(Jin-Young Kim) 한국스마트미디어학회 2020 스마트미디어저널 Vol.9 No.3
Out - of - distribution (OOD) samples are frequently encountered when deploying a classification model in plenty of real-world machine learning-based applications. Those samples are normally sampling far away from the training distribution, but many classifiers still assign them high reliability to belong to one of the training categories. In this study, we address the problem of removing OOD examples by estimating marginal density estimation using variational autoencoder (VAE). We also investigate other proper methods, such as temperature scaling, Gaussian discrimination analysis, and label smoothing. We use Chonnam National University (CNU) weeds dataset as the in - distribution dataset and CIFAR-10, CalTeach as the OOD datasets. Quantitative results show that the proposed framework can reject the OOD test samples with a suitable threshold.
유광현(Gwanghyun Yu),이재원(Jaewon Lee),보호앙트롱(Vo Hoang Trong),당탄부(Dang Thanh Vu),후이트완녁(Huy Toan Nguyen),이주환(JooHwan Lee),신도성(Dosung Shin),김진영(Jinyoung Kim) 한국정보기술학회 2019 한국정보기술학회논문지 Vol.17 No.12
Weeds are a major object which is very harmful to crops. To remove the weeds effectively, we have to classify them accurately and use herbicides. As computing technology has developed, image-based machine learning methods have been studied in this field, specially convolutional neural network(CNN) based models have shown good performance in public image dataset. However, CNN with numerous training parameters and high computational amount. Thus, it works under high hardware condition of expensive GPUs in real application. To solve these problems, in this paper, a hierarchical architecture based deep-learning model is proposed. The experimental results show that the proposed model successfully classify 21 species of the exotic weeds. That is, the model achieve 97.2612% accuracy with a small number of parameters. Our proposed model with a few parameters is expected to be applicable to actual application of network based classification services.