• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.115-124
• /
• 2022

In this paper, YOLO and EAST models are tested to analyze their performance in text area detecting for real-world and normal text images. The earl ier YOLO models which include YOLOv3 have been known to underperform in detecting text areas for given images, but the recently released YOLOv4 and YOLOv5 achieved promising performances to detect text area included in various images. Experimental results show that both of YOLO v4 and v5 models are expected to be widely used for text detection in the filed of scene text recognition in the future.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.05a
• /
• pp.431-434
• /
• 2020

본 논문에서는 악성코드탐지 기법으로 n-grams를 사용한 특징 추출을 통해 이미지 인식 분야에서 널리 쓰이는 Convolutional Neural Network로 학습하는 프레임워크를 제안한다. 윈도우즈 실행 파일의 PE 포맷에서 특징을 추출하여 6-grams 확률을 구하고 grayscale 을 통해 이미지로 변환한다. 이것을 기존에 연구된 탐지방법과 비교하여 우수함을 보인다. 학습에 사용된 데이터는 총 55,000개로 5-folds 교차검증을 하였으며 예측 정확도는 98.87%였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.05a
• /
• pp.285-288
• /
• 2003

기존의 침입탐지 시스템은 오용탐지모델이 널리 사용되고 있다. 이 모델은 낮은 오판율(False Alarm rates)을 가지고 있으나 새로운 공격에 대해 전문가시스템(Expert Systems)에 의한 규칙추가를 필요로 하고, 그 규칙과 완전히 매칭되는 시그너처만 공격으로 탐지하므로 변형된 공격을 탐지하지 못한다는 문제점을 가지고 있다. 본 논문에서는 이러한 문제점을 보완하기 위해 주성분분석(Principle Component Analysis ; 이하 PCA)과 시간지연신경망(Time Delay Neural Network ; 이하 TDNN)을 이용한 침입탐지 시스템을 제안한다. 패킷은 PCA를 이용하여 주성분을 결정하고 패킷이미지패턴으로 만든다. 이 연속된 패킷이미지패턴을 시간지연신경망의 학습패턴으로 사용한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.264-267
• /
• 2018

본 연구에서는 API 호출을 은닉할 수 있는 새로운 유형의 유저모드 기반 루트킷으로 Cuckoo Sandbox를 회피하는 기법과 이를 탐지하기 위한 연구를 한다. Cuckoo Sandbox의 행위 분석을 회피하기 위해 잠재적으로 출현 가능한 은닉된 코드 이미지 기반의 신종 루트킷 원리를 연구하고 탐지하기 위한 방안을 함께 연구한다. 네이티브 API 호출 코드 영역을 프로세스 공간에 직접 적재하여 네이티브 API를 호출하는 기법은 Cuckoo Sandbox에서 여전히 잠재적으로 행위 분석 회피가 가능하다. 본 연구에서는 은닉된 외부주소 호출 코드 영역의 탐지를 위해 프로세스의 가상메모리 공간에서 실행 가능한 페이지 영역을 탐색 후 코사인 유사도 분석으로 이미지 탐지 실험을 하였으며, 코드 영역이 맵핑된 정렬 단위의 4가지 실험 조건에서 평균 83.5% 유사도 탐지 결과를 확인하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.11a
• /
• pp.845-848
• /
• 2017

In this paper, we implemented a system to detect dangerous situations by recognizing the dangerous elements for infants by reading 2D images of children's houses, parks, playgrounds, and living rooms where infants are present through Faster R-CNN. We have implemented a detection model based on data that can be easily obtained from real life. Currently, machine learning is commercialized based on speech recognition and behavior data. However, this model can be applied to various service fields Respectively.

• The KIPS Transactions:PartB
• /
• v.16B no.4
• /
• pp.289-298
• /
• 2009

Active Contour Model, that is, Snake algorithm is effective for detection and tracking the objects. However, this algorithm has some drawbacks; numerous parameters must be designed(weighting factors, iteration steps, etc.), a reasonable initialization must be available and moreover suffers from numerical instability. Therefore we propose a novel Energy Corrected Snake(ECS) algorithm which improved on external energy of Snake algorithm for detection and tracking the moving object more effectively. The proposed algorithm uses the difference image, getting when the object is moving. It copies four direction images from the difference image and performs the accumulating compute to erasing image noise, so that it gets external energy steadily. Then external energy united with contour that is computed by internal energy. Consequently we can detect and track the moving object more speedily and easily. To show the effectiveness of the proposed algorithm, we experiment on 3 situations. The experimental results showed that the proposed algorithm outperformed by 6$\sim$9% of detection rate and 6$\sim$11% of tracker detection rate compared with the Snake algorithm.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.05a
• /
• pp.166-168
• /
• 2018

새롭게 변형되는 대규모 악성코드들을 신속하게 탐지하기 위하여 인공지능 딥러닝을 이용한 악성코드 탐지 기법을 제안한다. 대용량의 고차원 악성코드를 저차원의 이미지로 변환하고, 딥러닝 합성곱신경망(Convolution Neural Network)을 통해 이미지의 악성코드 패턴을 학습하고 분류하였다. 본 논문에서는 악성코드 분류 모델의 성능을 검증하기 위하여 악성코드 종류별 분류 실험과 악성코드와 정상코드 분류 실험을 실시하였고 각각 97.6%, 87%의 정확도로 악성코드를 구별해 내었다. 본 논문에서 제안한 악성코드 탐지 모델은 차원 축소를 통해 10,868개(200GB)의 대규모 데이터에 대하여 10분 이내의 학습시간이 소요되어 새로운 악성코드 학습 및 대용량 악성코드 탐지를 신속하게 처리 가능함을 보였다.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.2
• /
• pp.76-82
• /
• 2017

Malware, including ransomware to quickly detect, in this study, to provide an analysis method of malicious code through the image analysis that has been learned in the deep learning of artificial intelligence. First, to analyze the 2,400 malware data, and learning in artificial neural network Convolutional neural network and to image data. Extracts subgraphs to convert the graph of abstracted image, summarizes the set represent malware. The experimentally analyzed the malware is not how similar. Using deep learning of artificial intelligence by classifying malware and It shows the possibility of accurate malware detection.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.154-157
• /
• 2022

In the existing Tensorflow CNN environment, the object detection method is a method of performing object labeling and detection by Tensorflow itself. However, with the advent of YOLO, the efficiency of image object detection has increased. As a result, more deep layers can be built than existing neural networks, and the image object recognition rate can be increased. Therefore, in this paper, the detection ability and speed were compared and analyzed by designing an object detection system based on Darknet and YOLO and performing multi-layer construction and learning based on the existing convolutional neural network. For this reason, in this paper, a neural network methodology that efficiently uses Darknet's learning is presented.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.9
• /
• pp.49-55
• /
• 2021

In this paper, an enhanced 2-stage vehicle license plate detection scheme using object segmentation is proposed to detect accurately the rotated license plates due to the inclined photographing angles in real-road situations. With the previous 3-stage vehicle license plate detection pipeline model, the detection accuracy is likely decreased as the license plates are declined. To resolve this problem, we propose an enhanced 2-stage model by replacing the frontal two processing stages which are for detecting vehicle area and vehicle license plate respectively in only rectangular shapes in the previous 3-stage model with one step to detect vehicle license plate in arbitrarily shapes using object segmentation. According to the comparison results in terms of the detection accuracy of the proposed 2-stage scheme and the previous 3-stage pipeline model against the rotated license plates, the accuracy of the proposed 2-stage scheme is improved by up to about 20% even though the detection process is simplified.