• Annual Conference of KIPS
• /
• 2022.11a
• /
• pp.468-470
• /
• 2022

본 논문은 강화학습 POMDP(Partially Observable Markov Decision Process) 알고리즘을 사용하여 자갈밭과 같은 비평탄 지형을 극복하는 4족 보행 지능로봇을 설계하고 딥러닝 기법을 사용하여 사람을 검출한다. 로봇의 임베디드 환경에서 1단계 검출 알고리즘인 YOLO-v7과 SSD의 기본 모델, 경량 또는 네트워크 교체 모델의 성능을 비교하고 선정된 SSD MobileNet-v2의 검출 속도를 개선하기 위해 TensorRT를 사용하여 최적화를 진행하였다

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.221-222
• /
• 2022

본 논문에서는 모발 정밀검사(Phototrichogram)를 통해 일정 간격을 두고 촬영된 환자의 모발 두피 사진을 이용하여 머리카락 검출 및 머리카락의 개수 변화 추이에 따른 환자의 탈모 진단에 도움을 줄 방법을 제안하였다. 모발 정밀검사를 진행하여 촬영된 환자의 모발 사진으로부터 딥러닝 기반의 영상 분할 기법(Image Segmentation)의 하나인 DetectoRS 모델을 활용하여 머리카락을 자동 검출한다. 실험 결과 DetectoRS 모델의 분할 성능은 74.74%로 효과적으로 머리카락을 검출하였음을 확인할 수 있었다.

• KIISE Transactions on Computing Practices
• /
• v.23 no.4
• /
• pp.244-249
• /
• 2017

Steganalysis is to detect information hidden by steganography inside general data such as images. There are stegoanalysis techniques that use machine learning (ML). Existing ML approaches to steganalysis are based on extracting features from stego images and modeling them. Recently deep learning-based methodologies have shown significant improvements in detection accuracy. However, all the existing methods, including deep learning-based ones, have a critical limitation in that they can only detect stego images that are created by a specific steganography method. In this paper, we propose a generalized steganalysis method that can model multiple types of stego images using deep learning. Through various experiments, we confirm the effectiveness of our approach and envision directions for future research. In particular, we show that our method can detect each type of steganography with the same level of accuracy as that of a steganalysis method dedicated to that type of steganography, thereby demonstrating the general applicability of our approach to multiple types of stego images.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.5
• /
• pp.697-704
• /
• 2018

Since the face in image content corresponds to individual information that can distinguish a specific person from other people, it is important to accurately detect faces not hidden in an image. In this paper, we propose a method to accurately detect a face from input images using a deep learning algorithm, which is one of the machine learning methods. In the proposed method, image input via the red-green-blue (RGB) color model is first changed to the luminance-chroma: blue-chroma: red-chroma ($YC_bC_r$) color model; then, other regions are removed using the learned skin color model, and only the skin regions are segmented. A CNN model-based deep learning algorithm is then applied to robustly detect only the face region from the input image. Experimental results show that the proposed method more efficiently segments facial regions from input images. The proposed face area-detection method is expected to be useful in practical applications related to multimedia and shape recognition.

• Journal of Korea Society of Industrial Information Systems
• /
• v.8 no.1
• /
• pp.18-25
• /
• 2003

In this Paper, we propose a dissolve detection method based on video editing model. Our method consists of two steps In the first step, the candidate regions are found by using the first md second derivative of a variance curve. In a variance curve, a dissolve presents a parabola that is downward convex. Therefore the parabola is found as a candidate region for a dissolve. In the second step, the candidate region is verified for a dissolve region. In each candidate region, a variance at a valley of the parabola corresponding to dissolve is estimated and then the candidate region is verified by using estimated valley's variance. The valley's variance is determined by neighbor scene variances, so proposed method is adaptive to detect dissolve with various variances. Experiment results on video of various content types are reported and validated.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.5
• /
• pp.251-256
• /
• 2021

Unlike general cameras, a high-speed camera capable of capturing a large number of frames per second can enable the advancement of some image processing technologies that have been limited so far. This paper proposes a method of removing undesirable noise from an high-speed input color image, and then detecting a human face from the noise-free image. In this paper, noise pixels included in the ultrafast input image are first removed by applying a bidirectional filter. Then, using RetinaFace, a region representing the person's personal information is robustly detected from the image where noise was removed. The experimental results show that the described algorithm removes noise from the input image and then robustly detects a human face using the generated model. The model-based face-detection method presented in this paper is expected to be used as basic technology for many practical application fields related to image processing and pattern recognition, such as indoor and outdoor building monitoring, door opening and closing management, and mobile biometric authentication.

• Journal of Broadcast Engineering
• /
• v.26 no.2
• /
• pp.184-196
• /
• 2021

This paper proposes statistics adaptive linear regression-based object size prediction method for object detection. YOLOv2 and YOLOv3, which are typical deep learning-based object detection algorithms, designed the last layer of a network using statistics adaptive exponential regression model to predict the size of objects. However, an exponential regression model can propagate a high derivative of a loss function into all parameters in a network because of the property of an exponential function. We propose statistics adaptive linear regression layer to ease the gradient exploding problem of the exponential regression model. The proposed statistics adaptive linear regression model is used in the last layer of the network to predict the size of objects with statistics estimated from training dataset. We newly designed the network based on the YOLOv3tiny and it shows the higher performance compared to YOLOv3 tiny on the UFPR-ALPR dataset.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.4
• /
• pp.577-586
• /
• 2022

Detection and classification of steel surface defects are critical for product quality control in the steel industry. However, due to its low accuracy and slow speed, the traditional approach cannot be effectively used in a production line. The current, widely used algorithm (based on deep learning) has an accuracy problem, and there are still rooms for development. This paper proposes a method of steel surface defect detection combining EfficientNetV2 for image classification and YOLOv5 as an object detector. Shorter training time and high accuracy are advantages of this model. Firstly, the image input into EfficientNetV2 model classifies defect classes and predicts probability of having defects. If the probability of having a defect is less than 0.25, the algorithm directly recognizes that the sample has no defects. Otherwise, the samples are further input into YOLOv5 to accomplish the defect detection process on the metal surface. Experiments show that proposed model has good performance on the NEU dataset with an accuracy of 98.3%. Simultaneously, the average training speed is shorter than other models.

• Journal of Satellite, Information and Communications
• /
• v.12 no.1
• /
• pp.1-5
• /
• 2017

In this paper, multiuser detector (MUD) based on radial basis function (RBF) is proposed and simulated for a multicode direct sequence/code division multiple access (DS/CDMA) system in a multipath fading channel. The performance of RBF-based MUD is compared with that of many suboptimal multiuser detectors in terms of bit error probability. From the simulation results, it is confirmed that the RBF-based MUD outperforms decorrelating detector, and achieves near-optimum performance under various environments. The results in this paper can be applied to design of MUD for a multicode DS/CDMA system.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.460-462
• /
• 2021

Food basically must have nutrition and safety. Recently, a number of symptoms of food poisoning occurred in a kindergarten in Ansan, where food safety was suspected. Therefore, the safety of food is more demanding. In this paper, we propose a method to inprove the detector to secure food safety. The proposed method is to learn through the network of convolution neural network (CNN) and Faster region-CNN (Faster R-CNN) and test the images of normal and foreign products. As a result of testing through a deep learning model, the method that used Faster R-CNN in parallel with the existing foreign body detector algorithm showed better detection rate than other methods.