• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.18 no.4
• /
• pp.58-70
• /
• 2019

In the case of a pedestrian traffic accident, it has a large-scale danger directly connected by a fatal accident at the time of the accident. The domestic ITS is not used for intelligent risk classification because it is used only for collecting traffic information despite of the construction of good quality traffic infrastructure. The CNN based pedestrian detection classification model, which is a major component of the proposed system, is implemented on an embedded system assuming that it is installed and operated in a restricted environment. A new model was created by improving YOLO's artificial neural network, and the real-time detection speed result of average accuracy 86.29% and 21.1 fps was shown with 20,000 iterative learning. And we constructed a protocol interworking scenario and implementation of a system that can connect with the ITS. If a pedestrian accident prevention system connected with ITS will be implemented through this study, it will help to reduce the cost of constructing a new infrastructure and reduce the incidence of traffic accidents for pedestrians, and we can also reduce the cost for system monitoring.

• Journal of Internet Computing and Services
• /
• v.20 no.4
• /
• pp.91-102
• /
• 2019

In autonomous driving systems, the ability to classify pedestrians in images captured by cameras is very important for pedestrian safety. In the past, after extracting features of pedestrians with HOG(Histogram of Oriented Gradients) or SIFT(Scale-Invariant Feature Transform), people classified them using SVM(Support Vector Machine). However, extracting pedestrian characteristics in such a handcrafted manner has many limitations. Therefore, this paper proposes a method to classify pedestrians reliably and effectively using CNN's(Convolutional Neural Network) deep features and transfer learning. We have experimented with both the fixed feature extractor and the fine-tuning methods, which are two representative transfer learning techniques. Particularly, in the fine-tuning method, we have added a new scheme, called M-Fine(Modified Fine-tuning), which divideslayers into transferred parts and non-transferred parts in three different sizes, and adjusts weights only for layers belonging to non-transferred parts. Experiments on INRIA Person data set with five CNN models(VGGNet, DenseNet, Inception V3, Xception, and MobileNet) showed that CNN's deep features perform better than handcrafted features such as HOG and SIFT, and that the accuracy of Xception (threshold = 0.5) isthe highest at 99.61%. MobileNet, which achieved similar performance to Xception and learned 80% fewer parameters, was the best in terms of efficiency. Among the three transfer learning schemes tested above, the performance of the fine-tuning method was the best. The performance of the M-Fine method was comparable to or slightly lower than that of the fine-tuningmethod, but higher than that of the fixed feature extractor method.

• Journal of IKEEE
• /
• v.22 no.4
• /
• pp.1140-1146
• /
• 2018

Recently, a methodology for analyzing complex bio-signals using a deep learning model has emerged among studies that recognize human emotions. At this time, the accuracy of emotion classification may be changed depending on the evaluation method and reliability depending on the kind of data to be learned. In the case of biological signals, the reliability of data is determined according to the noise ratio, so that the noise detection method is as important as that. Also, according to the methodology for defining emotions, appropriate emotional evaluation methods will be needed. In this paper, we propose a wavelet -based noise threshold setting algorithm for verifying the reliability of data for multimodal bio-signal data labeled Valence and Arousal and a method for improving the emotion recognition rate by weighting the evaluation data. After extracting the wavelet component of the signal using the wavelet transform, the distortion and kurtosis of the component are obtained, the noise is detected at the threshold calculated by the hampel identifier, and the training data is selected considering the noise ratio of the original signal. In addition, weighting is applied to the overall evaluation of the emotion recognition rate using the euclidean distance from the median value of the Valence-Arousal plane when classifying emotional data. To verify the proposed algorithm, we use ASCERTAIN data set to observe the degree of emotion recognition rate improvement.

• The Journal of Bigdata
• /
• v.3 no.2
• /
• pp.19-33
• /
• 2018

Owing to the increase of FTA, food trade, and versatile preferences of consumers, food import has increased at tremendous rate every year. While the inspection check of imported food accounts for about 20% of the total food import, the budget and manpower necessary for the government's import inspection control is reaching its limit. The sudden import food accidents can cause enormous social and economic losses. Therefore, predictive system to forecast the compliance of food import with its preemptive measures will greatly improve the efficiency and effectiveness of import safety control management. There has already been a huge data accumulated from the past. The processed foods account for 75% of the total food import in the import food sector. The analysis of big data and the application of analytical techniques are also used to extract meaningful information from a large amount of data. Unfortunately, not many studies have been done regarding analyzing the import food and its implication with understanding the big data of food import. In this context, this study applied a variety of classification algorithms in the field of machine learning and suggested a data preprocessing method through the generation of new derivative variables to improve the accuracy of the model. In addition, the present study compared the performance of the predictive classification algorithms with the general base classifier. The Gaussian Naïve Bayes prediction model among various base classifiers showed the best performance to detect and predict the nonconformity of imported food. In the future, it is expected that the application of the abnormality detection model using the Gaussian Naïve Bayes. The predictive model will reduce the burdens of the inspection of import food and increase the non-conformity rate, which will have a great effect on the efficiency of the food import safety control and the speed of import customs clearance.

• Phonetics and Speech Sciences
• /
• v.12 no.4
• /
• pp.91-98
• /
• 2020

This study investigated the acoustic characteristics of sustained vowel /a/ and sentence utterance produced by patients with muscle tension dysphonia (MTD) using cepstrum-based acoustic variables. 36 women diagnosed with MTD and the same number of women with normal voice participated in the study and the data were recorded and measured by ADSVTM. The results demonstrated that cepstral peak prominence (CPP) and CPP_F0 among all of the variables were statistically significantly lower than those of control group. When it comes to the GRBAS scale, overall severity (G) was most prominent, and roughness (R), breathiness (B), and strain (S) indices followed in order in the voice quality of MTD patients. As these characteristics increased, a statistically significant negative correlation was observed in CPP. We tried to classify MTD and control group using CPP and CPP_F0 variables. As a result of statistic modeling with a Random Forest machine learning algorithm, much higher classification accuracy (100% in training data and 83.3% in test data) was found in the sentence reading task, with CPP being proved to be playing a more crucial role in both vowel and sentence reading tasks.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.1
• /
• pp.1-8
• /
• 2021

In the political field of news articles, there are polarized and biased characteristics such as conservative and liberal, which is called political bias. We constructed keyword-based dataset to classify bias of news articles. Most embedding researches represent a sentence with sequence of morphemes. In our work, we expect that the number of unknown tokens will be reduced if the sentences are constituted by subwords that are segmented by the language model. We propose a document embedding model with subword tokenization and apply this model to SVM and feedforward neural network structure to classify the political bias. As a result of comparing the performance of the document embedding model with morphological analysis, the document embedding model with subwords showed the highest accuracy at 78.22%. It was confirmed that the number of unknown tokens was reduced by subword tokenization. Using the best performance embedding model in our bias classification task, we extract the keywords based on politicians. The bias of keywords was verified by the average similarity with the vector of politicians from each political tendency.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.133-140
• /
• 2022

We investigate the performance of deep learning-based Korean language models on a task of predicting the score range of Korean essays written by foreign students. We construct a data set containing a total of 304 essays, which include essays discussing the criteria for choosing a job ('job'), conditions of a happy life ('happ'), relationship between money and happiness ('econ'), and definition of success ('succ'). These essays were labeled according to four letter grades (A, B, C, and D), and a total of eleven essay score range prediction experiments were conducted (i.e., five for predicting the score range of 'job' essays, five for predicting the score range of 'happiness' essays, and one for predicting the score range of mixed topic essays). Three deep learning-based Korean language models, KoBERT, KcBERT, and KR-BERT, were fine-tuned using various training data. Moreover, two traditional probabilistic machine learning classifiers, naive Bayes and logistic regression, were also evaluated. Experiment results show that deep learning-based Korean language models performed better than the two traditional classifiers, with KR-BERT performing the best with 55.83% overall average prediction accuracy. A close second was KcBERT (55.77%) followed by KoBERT (54.91%). The performances of naive Bayes and logistic regression classifiers were 52.52% and 50.28% respectively. Due to the scarcity of training data and the imbalance in class distribution, the overall prediction performance was not high for all classifiers. Moreover, the classifiers' vocabulary did not explicitly capture the error features that were helpful in correctly grading the Korean essay. By overcoming these two limitations, we expect the score range prediction performance to improve.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.101-125
• /
• 2022

Recently, with the development of computing technology and the improvement of the cloud environment, deep learning technology has developed, and attempts to apply deep learning to various fields are increasing. A typical example is anomaly detection, which is a technique for identifying values or patterns that deviate from normal data. Among the representative types of anomaly detection, it is very difficult to detect a contextual anomaly that requires understanding of the overall situation. In general, detection of anomalies in image data is performed using a pre-trained model trained on large data. However, since this pre-trained model was created by focusing on object classification of images, there is a limit to be applied to anomaly detection that needs to understand complex situations created by various objects. Therefore, in this study, we newly propose a two-step pre-trained model for detecting abnormal situation. Our methodology performs additional learning from image captioning to understand not only mere objects but also the complicated situation created by them. Specifically, the proposed methodology transfers knowledge of the pre-trained model that has learned object classification with ImageNet data to the image captioning model, and uses the caption that describes the situation represented by the image. Afterwards, the weight obtained by learning the situational characteristics through images and captions is extracted and fine-tuning is performed to generate an anomaly detection model. To evaluate the performance of the proposed methodology, an anomaly detection experiment was performed on 400 situational images and the experimental results showed that the proposed methodology was superior in terms of anomaly detection accuracy and F1-score compared to the existing traditional pre-trained model.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.40 no.5
• /
• pp.367-380
• /
• 2022

In this study, the periodic and simplified update and production way of the ECVAM (Environmental Conservation Value Assessment Map) was presented through the classification of environmental values using KOMPSAT-3A satellite imagery and land cover map. ECVAM is a map that evaluates the environmental value of the country in five stages based on 62 legal evaluation items and 8 environmental and ecological evaluation items, and is provided on two scales: 1:25000 and 1:5000. However, the 1:5000 scale environmental assessment map is being produced and serviced with a slow renewal cycle of one year due to various constraints such as the absence of reference materials and different production years. Therefore, in this study, one of the deep learning techniques, KOMPSAT-3A satellite image, SI (Spectral Indices), and land cover map were used to conduct this study to confirm the possibility of establishing an environmental assessment map. As a result, the accuracy was calculated to be 87.25% and 85.88%, respectively. Through the results of the study, it was possible to confirm the possibility of constructing an environmental assessment map using satellite imagery, optical index, and land cover classification.

• Economic and Environmental Geology
• /
• v.55 no.4
• /
• pp.389-398
• /
• 2022

Many studies have been conducted to accurately predict the correlations between As and heavy metals content in contaminated soil and cultivated crops; however, due to the low correlation between the two, few clear results were obtained to date. This study aimed to create statistical models that predict the As content transferred from soil to polished rice, considering the physicochemical properties of the soil, as well as the total content and the single-extracted content of As in the soil. Predictive models were derived through regression analysis while sequentially classifying soil samples according to pH, soluble As content by single extraction, and organic matter content of the soil. The correlation coefficients between the As content in 80 polished rice and total As content and Mehlich soluble As content in the soil were low, 0.533 and 0.493, respectively. However, the models derived after sequential classification of the soil by pH, a ratio of total As content to Mehlich soluble As content, and organic matter content greatly increased the predictive power; ① 0.963 for 13 soils with a pH higher than 6.5, ② 0.849 for 15 soils with pH lower than 6.5 and a high ratio of As_Tot/As_Mehlich, ③ 0.935 for 30 soils with pH lower than 6.5, a high ratio of As_Tot/As_Mehlich, and organic matter content lower than 8.5%. The suggested prediction model of As transfer from soil to polished rice derived by soil classification may serve as a statistically significant methodology in establishing a rice cultivation standard for arsenic-contaminated soil.