• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.10
• /
• pp.323-328
• /
• 2020

Recently, new approaches that significantly improve performance in object detection and recognition using deep learning technology have been proposed quickly. Of the various techniques for object detection, especially facial object detection (Faster R-CNN, R-CNN, YOLO, SSD, etc), SSD is superior in accuracy and speed to other techniques. At the same time, multiple object detection networks are also readily available. In this paper, among object detection networks, Mobilenet v2 network is used, models combined with SSDs are trained, and methods for detecting objects at a rate of four times or more than conventional performance are proposed using TensorRT engine, and the performance is verified through experiments. Facial object detector was created as an application to verify the performance of the proposed method, and its behavior and performance were tested in various situations.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.05a
• /
• pp.574-576
• /
• 2020

최근에는 딥러닝 기술의 발달로 물체 인식 및 검출에 관한 기술들 또한 발탄하고 있다. 검출에 관한 여러 기법(Faster R-CNN, R-CNN, YOLO, SSD 등) 중 SSD는 다른 기법들과는 다르게 높은 정확도와 빠른 속도가 특징이다. 동시에 여러 detection network들도 쉽게 이용이 가능하다. 본 논문에서는 detection netowork중 Mobilenet V2 network를 이용하여 SSD와 결합해 모델을 훈련하고, TensorRT engine을 이용하여 더 빠른 속도로 검출할 수 있는 방법에 대해 논의한다. 이 방법을 통해 face detector를 만들어 여러 상황에서 쓰일 수 있도록 한다.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.5
• /
• pp.467-474
• /
• 2019

In this paper, we propose a novel method to improve FPS while maintaining the accuracy of YOLO v2 model in NVIDIA Jetson TX1. In general, in order to reduce the amount of computation, a conversion to an integer operation or reducing the depth of a network have been used. However, the accuracy of recognition can be deteriorated. So, we use methods to reduce computation and memory consumption through adjustment of the filter size and integrated computation of the network The first method is to replace the $3{\times}3$ filter with a $1{\times}1$ filter, which reduces the number of parameters to one-ninth. The second method is to reduce the amount of computation through CBR (Convolution-Add Bias-Relu) among the inference acceleration functions of TensorRT, and the last method is to reduce memory consumption by integrating repeated layers using TensorRT. For the simulation results, although the accuracy is decreased by 1% compared to the existing YOLO v2 model, the FPS has been improved from the existing 3.9 FPS to 11 FPS.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.154-155
• /
• 2023

오픈 소스인 YOLO(You Only Look Once) 객체 탐지 알고리즘이 공개된 이후, 산업 현장에서는 고성능 컴퓨터에서 벗어나 효율과 특수한 환경에 사용하기 위해 임베디드 시스템에 도입하고 있다. 그러나, NVIDIA의 Jetson nano의 경우, Pytorch의 YOLOv7 딥러닝 모델에 대한 추론이 진행되지 않는다. 따라서 제한적인 전력과 메모리, 연산능력 최적화 과정은 필수적이다. 본 논문은 NVIDIA의 임베디드 플랫폼 Jetson 계열의 Xavier NX, Orin AGX, Nano에서 딥러닝 모델을 적용하기 위한 최적화 과정과 플랫폼에서 다양한 크기의 YOLOv7의 PyTorch 모델들을 Tensor RT로 변환하여 FPS(Frames Per Second)를 측정 및 비교한다. 측정 결과를 통해, 각 임베디드 플랫폼에서 YOLOv7 모델의 추론은 Tensor RT는 Pytorch에서 약 4.1배 적은 FPS 변동성과 약 2.25배 정도의 FPS 속도향상을 보였다.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.15 no.2
• /
• pp.51-60
• /
• 2020

Optical character recognition (OCR) has been studied for decades because it is very useful in a variety of places. Nowadays, OCR's performance has improved significantly due to outstanding deep learning technology. Thus, there is an increasing demand for commercial-grade but affordable OCR systems. We have developed a low-cost, high-performance OCR system for the industry with the cheapest embedded developer kit that supports GPU acceleration. To achieve high accuracy for industrial use on limited computing resources, we chose a state-of-the-art text recognition algorithm that uses an end-to-end deep learning network as a baseline model. The model was then improved by replacing the feature extraction network with the best one suited to our conditions. Among the various candidate networks, EfficientNet-B3 has shown the best performance: excellent recognition accuracy with relatively low memory consumption. Besides, we have optimized the model written in TensorFlow's Python API using TensorFlow-TensorRT integration and TensorFlow's C++ API, respectively.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.15 no.6
• /
• pp.281-287
• /
• 2020

In this paper, we propose a method for estimating a walking direction by which a mobile robots follows a person using TRT (Tensor RT) pose, which is motion recognition based on deep learning. Mobile robots can measure individual movements by recognizing key points on the person's pelvis and determine the direction in which the person tries to move. Using these information and the distance between robot and human, the mobile robot can follow the person stably keeping a safe distance from people. The TRT Pose only extracts key point information to prevent privacy issues while a camera in the mobile robot records video. To validate the proposed technology, experiment is carried out successfully where human walks away or toward the mobile robot in zigzag form and the robot continuously follows human with prescribed distance.

• Proceedings of the Korea Information Processing Society Conference
• /
• 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 Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.408-410
• /
• 2022

최근 데이터 획득 위치에 가장 근접하고, 저 수준의 계산력을 제공하는 엣지 기기를 중심으로 직접 딥러닝 추론을 수행하고자 하는 요구가 증가하고 있다. 본 논문에서는 드론에서 촬영한 교통 영상 데이터를 기반으로, 다수의 차량 종류 및 보행자를 식별하는 모델을 Jetson Nano 에 탑재하여 기본 성능을 측정한다. 더불어, 자원제약형 기기 환경에서 TensorRT 와 Deepstream 을 활용하여 객체 식별 모델의 연산 경량화 및 추론 가속화 성능을 극대화하기 위한 구현 및 실험을 수행하여 Anchor-based 및 Anchor-free 객체 식별 모델의 정확도와 실시간 대응력을 평가하고 논의한다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.4
• /
• pp.49-57
• /
• 2022

The government is making great efforts to prevent crimes such as missing children by using public CCTVs. However, there is a shortage of operating manpower, weakening of concentration due to long-term concentration, and difficulty in tracking. In addition, applying real-time object search, re-identification, and tracking through a deep learning algorithm showed a phenomenon of increased parameters and insufficient memory for speed reduction due to complex network analysis. In this paper, we designed the network to improve speed and save memory through the application of Yolo v4, which can recognize real-time objects, and the application of Batch and TensorRT technology. In this thesis, based on the research on these advanced algorithms, OSNet re-ranking and K-reciprocal nearest neighbor for re-identification, Jaccard distance dissimilarity measurement algorithm for correlation, etc. are developed and used in the solution of CCTV national safety identification and tracking system. As a result, we propose a solution that can track objects by recognizing and re-identification objects in real-time within situation of a Korean public multi-CCTV environment through a set of algorithm combinations.

• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.4
• /
• pp.1-9
• /
• 2023

In this paper, we propose a method to recognize the braille blocks for embedded devices in real time through deep learning. First, a deep learning model for braille block recognition is trained on a high-performance computer, and the learning model is applied to a lightweight tool to apply to an embedded device. To recognize the walking information of the braille block, an algorithm is used to determine the path using the distance from the braille block in the image. After detecting braille blocks, bollards, and crosswalks through the YOLOv8 model in the video captured by the embedded device, the walking information is recognized through the braille block path discrimination algorithm. We apply the model lightweight tool to YOLOv8 to detect braille blocks in real time. The precision of YOLOv8 model weights is lowered from the existing 32 bits to 8 bits, and the model is optimized by applying the TensorRT optimization engine. As the result of comparing the lightweight model through the proposed method with the existing model, the path recognition accuracy is 99.05%, which is almost the same as the existing model, but the recognition speed is reduced by 59% compared to the existing model, processing about 15 frames per second.