• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.75-77
• /
• 2021

전력 생산의 효율을 높이고 지속적인 공정관리를 위해 전력 설비 데이터의 패턴을 분석하고 원인이 되는 주요 변수를 찾는 것이 중요하다. 따라서, 본 연구에서는 전력 설비 데이터의 패턴을 분석하기 위해 데이터를 군집화하고 연구 방법으로 Decision Tree, Random Forest와 ResNet을 이용하여 패턴을 분류하였다. Class Activation Map을 이용하여 설비데이터의 원인이 되는 주요 변수를 확인하였다. 본 연구를 통해 전력 설비 데이터의 분류 및 원인 분석이 가능한 통합적 솔루션을 제시하고자 한다.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.287-292
• /
• 2019

다양한 매체를 통해 텍스트 데이터가 빠르게 생성되면서 요약된 텍스트에 대한 수요가 증가하고 있다. 시퀀스-투-시퀀스 모델의 등장과 attention 기법의 출현은 추상적 요약의 난도를 낮추고 성능을 상승시켰다. 그러나 그동안 진행되어 온 attention 기반의 시퀀스-투-시퀀스 모델을 통한 요약 관련 연구들은 요약 시 텍스트의 카테고리 정보를 이용하지 않았다. 텍스트의 카테고리 정보는 Class Activation Map(CAM)을 통해 얻을 수 있는데, 텍스트를 요약할 때 핵심이 되는 단어와 CAM에서 높은 수치를 보이는 단어가 상당수 일치한다는 사실은 요약문 생성이 텍스트의 카테고리에 의존적일 필요가 있음을 증명한다. 본 논문에서는 요약문 생성 시 집중 정도에 대한 정보를 CAM을 통해 전달하여 attention matrix를 보강할 수 있는 모델을 제안하였다. 해당 모델을 사용하여 요약문을 생성하고 대표적인 요약 성능 지표인 ROUGE로 측정한 결과, attention 기반의 시퀀스-투-시퀀스 모델이 질이 떨어지는 요약문을 생성할 때 attention의 성능을 보강하여 요약문의 질을 높일 수 있음을 알 수 있었다.

• Nuclear Engineering and Technology
• /
• v.54 no.12
• /
• pp.4684-4692
• /
• 2022

Radionuclide identification is an important part of the nuclear material identification system. The development of artificial intelligence and machine learning has made nuclide identification rapid and automatic. However, many methods directly use existing deep learning models to analyze the gamma-ray spectrum, which lacks interpretability for researchers. This study proposes an explainable radionuclide identification algorithm based on the convolutional neural network and class activation mapping. This method shows the area of interest of the neural network on the gamma-ray spectrum by generating a class activation map. We analyzed the class activation map of the gamma-ray spectrum of different types, different gross counts, and different signal-to-noise ratios. The results show that the convolutional neural network attempted to learn the relationship between the input gamma-ray spectrum and the nuclide type, and could identify the nuclide based on the photoelectric peak and Compton edge. Furthermore, the results explain why the neural network could identify gamma-ray spectra with low counts and low signal-to-noise ratios. Thus, the findings improve researchers' confidence in the ability of neural networks to identify nuclides and promote the application of artificial intelligence methods in the field of nuclide identification.

• Korean Journal of Agricultural Science
• /
• v.50 no.3
• /
• pp.399-406
• /
• 2023

In this study, we propose a ripe tomato bunch localization method based on convolutional neural networks, to be applied in robotic harvesting systems. Tomato images were obtained from a smart greenhouse at the Rural Development Administration (RDA). The sample images for training were extracted based on tomato maturity and resized to 128 × 128 pixels for use in the classification model. The model was constructed based on four-layer convolutional neural networks, and the classes were determined based on stage of maturity, using a Softmax classifier. The localization of the ripe tomato bunch region was indicated on a class activation map. The class activation map could show the approximate location of the tomato bunch but tends to present a local part or a large part of the ripe tomato bunch region, which could lead to poor performance. Therefore, we suggest a recursive method to improve the performance of the model. The classification results indicated that the accuracy, precision, recall, and F1-score were 0.98, 0.87, 0.98, and 0.92, respectively. The localization performance was 0.52, estimated by the Intersection over Union (IoU), and through input recursion, the IoU was improved by 13%. Based on the results, the proposed localization of the ripe tomato bunch area can be incorporated in robotic harvesting systems to establish the optimal harvesting paths.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.9
• /
• pp.349-358
• /
• 2021

Recently, a method of finding the attention area or localization area for an object of an image using CAM (Class Activation Map)[1] has been variously carried out as a study of WSOL (Weakly Supervised Object Localization). The attention area extraction from the object heat map using CAM has a disadvantage in that it cannot find the entire area of the object by focusing mainly on the part where the features are most concentrated in the object. To improve this, using CAM and Selective Search[6] together, we first expand the attention area in the heat map, and a Gaussian smoothing is applied to the extended area to generate retraining data. Finally we train the data to expand the attention area of the objects. The proposed method requires retraining only once, and the search time to find an localization area is greatly reduced since the selective search is not needed in this stage. Through the experiment, the attention area was expanded from the existing CAM heat maps, and in the calculation of IOU (Intersection of Union) with the ground truth for the bounding box of the expanded attention area, about 58% was improved compared to the existing CAM.

• Journal of radiological science and technology
• /
• v.45 no.1
• /
• pp.41-47
• /
• 2022

The purpose of this study is to apply a deep learning model that can distinguish lung perfusion and lung ventilation images in nuclear medicine, and to evaluate the image classification ability. Image data pre-processing was performed in the following order: image matrix size adjustment, min-max normalization, image center position adjustment, train/validation/test data set classification, and data augmentation. The convolutional neural network(CNN) structures of VGG-16, ResNet-18, Inception-ResNet-v2, and SE-ResNeXt-101 were used. For classification model evaluation, performance evaluation index of classification model, class activation map(CAM), and statistical image evaluation method were applied. As for the performance evaluation index of the classification model, SE-ResNeXt-101 and Inception-ResNet-v2 showed the highest performance with the same results. As a result of CAM, cardiac and right lung regions were highly activated in lung perfusion, and upper lung and neck regions were highly activated in lung ventilation. Statistical image evaluation showed a meaningful difference between SE-ResNeXt-101 and Inception-ResNet-v2. As a result of the study, the applicability of the CNN model for lung scintigraphy classification was confirmed. In the future, it is expected that it will be used as basic data for research on new artificial intelligence models and will help stable image management in clinical practice.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.5
• /
• pp.55-67
• /
• 2022

Sensor data can provide fault diagnosis for equipment. However, the cause analysis for fault results of equipment is not often provided. In this study, we propose an explainable convolutional neural network framework for the sensor-based time series classification model. We used sensor-based time series dataset, acquired from vehicles equipped with sensors, and the Wafer dataset, acquired from manufacturing process. Moreover, we used Cycle Signal dataset, acquired from real world mechanical equipment, and for Data augmentation methods, scaling and jittering were used to train our deep learning models. In addition, our proposed classification models are convolutional neural network based models, FCN, 1D-CNN, and ResNet, to compare evaluations for each model. Our experimental results show that the ResNet provides promising results in the context of time series classification with accuracy and F1 Score reaching 95%, improved by 3% compared to the previous study. Furthermore, we propose XAI methods, Class Activation Map and Layer Visualization, to interpret the experiment result. XAI methods can visualize the time series interval that shows important factors for sensor data classification.

• Journal of the Korean Data and Information Science Society
• /
• v.18 no.3
• /
• pp.721-733
• /
• 2007

The purpose of this research was to determine the effects of the functional lateralization of the brain variables related to the sex, the scientific attitude and the scientific exploration skills. The science instruction is divided in each type of the lecturing class with the experiment class. As for the degree of SMR brainwave activation in each stage are presented while accumulating the brain waves from the right, left and the whole brain waves are analyzed during the science learning activities. It is therefore reasonable to consider the science instruction types and brain lateralization to enhance the science learning effectiveness. Sensorimotor rhythm brainwave as the low Beta is represented well to show the thought process. Category quantification scores and objective scores are calculated to show the visual positioning map for the relationships of the categories by homogeneity analysis.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.11
• /
• pp.439-448
• /
• 2021

Driver negligence accounts for the largest proportion of the causes of traffic accidents, and research to detect them is continuously being conducted. This paper proposes a method to accurately detect a distracted driver and localize the most characteristic parts of the driver. The proposed method hierarchically constructs a CNN basic model that classifies 10 classes based on CAM in order to detect driver distration and 4 subclass models for detailed classification of classes having a confusing or common feature area in this model. The classification result output from each model can be considered as a new feature indicating the degree of matching with the CNN feature maps, and the accuracy of classification is improved by horizontally combining and learning them. In addition, by combining the heat map results reflecting the classification results of the basic and detailed classification models, the characteristic areas of attention in the image are found. The proposed method obtained an accuracy of 95.14% in an experiment using the State Farm data set, which is 2.94% higher than the 92.2%, which is the highest accuracy among the results using this data set. Also, it was confirmed by the experiment that more meaningful and accurate attention areas were found than the results of the attention area found when only the basic model was used.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.5
• /
• pp.2558-2573
• /
• 2019

We present a deep CNN-based approach for classifying artistic media from artwork images. We aim to classify most frequently used artistic media including oilpaint brush, watercolor brush, pencil and pastel, etc. For this purpose, we extend VGGNet, one of the most widely used CNN structure, by substituting its last layer with a fully convolutional layer, which reveals class activation map (CAM), the region of classification. We build two artwork image datasets: YMSet that collects more than 4K artwork images for four most frequently used artistic media from various internet websites and WikiSet that collects almost 9K artwork images for ten most frequently used media from WikiArt. We execute a human baseline experiment to compare the classification performance. Through our experiments, we conclude that our classifier is superior in classifying artistic media to human.