http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
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.
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.
열화상 비디오에서 합성곱 신경망 기반의 실시간 인간 탐지
나지르 샤히드(Nazeer Shahid),유광현(Gwang-Hyun Yu),황성민(Seong-Min Hwang),보 호앙 트롱(Vo Hoang Trong),김진영(Jin-Young Kim) 대한전자공학회 2018 대한전자공학회 학술대회 Vol.2018 No.11
In this paper, we have proposed a Convolution Neural Network based human classification technique that efficiently operates in real time. Background subtraction is done using improved Running Gaussian Average to get the initial background model. Background updating is implemented using selectivity updating and random selection of background pixel from every new frame. Morphology is applied to extract ROIs from each frame. For classification, CNN model is trained and tested with our own dataset. For incorporating the model with real-time application, we neglect the nodes from computational graph that have no weights and convert other weights to constants. With this trained CNN model, ROI is classified as human or non-human in real-time. The processing time depends on number of ROI present in the frame. For our testing data, average processing time is 25fps.