• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.1
• /
• pp.144-150
• /
• 2017

Recently, deep learning is used for intelligent processing and accuracy improvement of data. It is formed calculation model composed of multi data processing layer that train the data representation through an abstraction of the various levels. A category of deep learning, convolution neural network is utilized in various research fields, which are human pose estimation, face recognition, image classification, speech recognition. When using the deep layer and lots of class, CNN that show a good performance on image classification obtain higher classification rate but occur the overfitting problem, when using a few data. So, we design the training network based on convolution neural network and trained our image data set for object classification in few class problem. The experiment show the higher classification rate of 7.06% in average than the previous networks designed to classify the object in 1000 class problem.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2024.01a
• /
• pp.17-18
• /
• 2024

악성 흉수의 진단은 세포학적 검사로 암세포를 확인하는 것이 필수적이며 진단율은 50~80%로 나타난다. 양성자 단층 촬영은 비침습적으로 암 병기를 평가하는 유용한 방법이다. 하지만 암이 아닌 다른 원인으로 인한 포도당 대사로 인하여 양전자 단층 촬영만으로 악성 흉수를 진단하는 데 어려움이 있다. 악성 흉수 자동 진단 모델은 암세포를 진단하는데 있어서 보조적인 역할이 가능하다. 이에 따라 본 연구는 컨볼루션 신경망 기반의 딥러닝 모델을 개발하여 악성 흉수 진단 성능을 확인하고 진단의 보조적 목적으로써 딥러닝의 사용 가능성을 확인하고자 하였다. 결과적으로 모델 전반적으로 accuracy 0.7~0.86의 높은 성능을 보였다. 본 연구의 결과를 통해 실제 의료 환경에서 악성 흉수를 진단하는데 딥러닝 모델이 보조적인 역할을 할 수 있을 것으로 기대된다.

• Annual Conference of KIPS
• /
• 2017.04a
• /
• pp.1031-1034
• /
• 2017

기존 번호판 검출 및 인식에 사용되어지는 방법은 사랑이 직접 번호판의 특정을 기술하여 검출을 진행한다. 본 연구에서는 학습 기반의 완전 컨볼루션 신경망을 이용하여 번호판을 검출하였고 신경망은 약 27MB의 용량만으로 110-FPS 정도의 성능을 얻었다. 학습을 위한 데이터는 한국 번호판의 모든 종류 및 주간, 야간의 환경을 포함한 대략 5000개를 직접 수집하였다 또한 5000개의 데이터를 회전 및 이동에 대한 무작위적인 변형을 주어 대략 15000개의 데이터로 확장하였다 확장된 데이터로 얻은 결과로 번호판 검출률 97%를 얻었다.

• Annual Conference of KIPS
• /
• 2016.04a
• /
• pp.754-757
• /
• 2016

기존 번호판 인식은 직접 특징 추출 알고리즘을 개발하여 완전 연결 신경망으로 특징을 분류하는 방법이 보편적이다. 본 연구는 전처리 과정에서 번호판 후보군 검출 및 세그먼테이션을 수행하고 특징 추출 없이 미리 학습된 심층 컨볼루션 신경망을 통해 문자를 분류하는 방법을 제안한다. 직접 수집한 2,900장의 번호판 데이터베이스를 이용하여 훈련 집합 및 검증 집합을 구성하였다. 훈련 집합과 검증 집합에 대해 실험한 결과 번호판 후보군 검출률은 97%를 얻을 수 있었고, 이에 대한 인식률은 95%를 얻었다.

• Annual Conference of KIPS
• /
• 2018.05a
• /
• pp.285-288
• /
• 2018

데이터 정형화기술은 자연어 처리 및 인공지능분야, 데이터베이스 등 다양한 분야에서 중요한 핵심적인 기술 중 하나이다. 최근 정형화 문제를 푸는 많은 신경망 기반 알고리즘들이 제안되었으나, 기존의 모든 알고리즘이 키워드의 후보가 입력으로 주어진다고 가정하고 있으며, 알고리즘 대부분은 두 개의 속성(attribute)을 가지는 이진 관계(binary relation)만 처리할 수 있다는 한계가 있다. 본 논문에서는 컨볼루션 신경망을 이용한 N항 관계 정형화 방업을 제안하고, 이를 이용한 구인 광고 정형화 시스템을 개발하고 성능을 평가한다.

• KIISE Transactions on Computing Practices
• /
• v.22 no.11
• /
• pp.614-618
• /
• 2016

As information on the internet and the data from smart devices are growing, the amount of stream data is also increasing in the real world. The stream data, which is a potentially large data, requires online learnable models and algorithms. In this paper, we propose a novel class of models: predictive convolutional neural networks to be able to perform online learning. These models are designed to deal with longer patterns as the layers become higher due to layering convolutional operations: detection and max-pooling on the time axis. As a preliminary check of the concept, we chose two-month gathered GPS data sequence as an observation sequence. On learning them with the proposed method, we compared the original sequence and the regenerated sequence from the abstract information of the models. The result shows that the models can encode long-range patterns, and can generate a raw observation sequence within a low error.

• The Journal of the Acoustical Society of Korea
• /
• v.37 no.4
• /
• pp.256-261
• /
• 2018

There are three types of noise generated inside the vehicle: BSR (Buzz, Squeak, Rattle). In this paper, we propose a classifier that automatically classifies automotive BSR noise by using features extracted from deep convolutional neural networks. In the preprocessing process, the features of above three noises are represented as noise-map using STFT (Short-time Fourier Transform) algorithm. In order to cope with the problem that the position of the actual noise is unknown in the part of the generated noise map, the noise map is divided using the sliding window method. In this paper, internal parameter of the deep convolutional neural networks is visualized using the t-SNE (t-Stochastic Neighbor Embedding) algorithm, and the misclassified data is analyzed in a qualitative way. In order to analyze the classified data, the similarity of the noise type was quantified by SSIM (Structural Similarity Index) value, and it was found that the retractor tremble sound is most similar to the normal travel sound. The classifier of the proposed method compared with other classifiers of machine learning method recorded the highest classification accuracy (99.15 %).

• Journal of Korea Society of Industrial Information Systems
• /
• v.27 no.2
• /
• pp.25-34
• /
• 2022

With the development of online services, the ripple effect of negative comments is increasing, and the damage of cyber violence is rising. Various methods such as filtering based on forbidden words and reporting systems prevent this, but it is challenging to eradicate negative comments. Therefore, this study aimed to increase the accuracy of the classification of negative comments using deep learning and blur the parts corresponding to profanity. Two different conditional training helped decide the number of deep learning layers and filters. The accuracy of 88% confirmed with 90% of the dataset for training and 10% for tests. In addition, Grad-CAM enabled us to find and blur the location of swear words in negative comments. Although the accuracy of classifying comments based on simple forbidden words was 56%, it was found that blurring negative comments through the deep learning model was more effective.

• Communications of the Korean Institute of Information Scientists and Engineers
• /
• v.36 no.2
• /
• pp.25-31
• /
• 2018

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.3
• /
• pp.123-128
• /
• 2019

This paper proposes the stock price prediction based on the artificial intelligence, where the model with recurrent convolution neural network (RCNN) layers is adopted. In the motivation of this prediction, long short-term memory model (LSTM)-based neural network can make the output of the time series prediction. On the other hand, the convolution neural network provides the data filtering, averaging, and augmentation. By combining the advantages mentioned above, the proposed technique predicts the estimated stock price of next day. In addition, in order to emphasize the recent time series, a custom weighted loss function is adopted. Moreover, stock data related to the stock price index are adopted to consider the market trends. In the experiments, the proposed stock price prediction reduces the test error by 3.19%, which is over other techniques by about 19%.