• 산업경영시스템학회지
• /
• 제46권1호
• /
• pp.55-67
• /
• 2023

Lightweight face recognition models, as one of the most popular and long-standing topics in the field of computer vision, has achieved vigorous development and has been widely used in many real-world applications due to fewer number of parameters, lower floating-point operations, and smaller model size. However, few surveys reviewed lightweight models and reimplemented these lightweight models by using the same calculating resource and training dataset. In this survey article, we present a comprehensive review about the recent research advances on the end-to-end efficient lightweight face recognition models and reimplement several of the most popular models. To start with, we introduce the overview of face recognition with lightweight models. Then, based on the construction of models, we categorize the lightweight models into: (1) artificially designing lightweight FR models, (2) pruned models to face recognition, (3) efficient automatic neural network architecture design based on neural architecture searching, (4) Knowledge distillation and (5) low-rank decomposition. As an example, we also introduce the SqueezeFaceNet and EfficientFaceNet by pruning SqueezeNet and EfficientNet. Additionally, we reimplement and present a detailed performance comparison of different lightweight models on the nine different test benchmarks. At last, the challenges and future works are provided. There are three main contributions in our survey: firstly, the categorized lightweight models can be conveniently identified so that we can explore new lightweight models for face recognition; secondly, the comprehensive performance comparisons are carried out so that ones can choose models when a state-of-the-art end-to-end face recognition system is deployed on mobile devices; thirdly, the challenges and future trends are stated to inspire our future works.

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 1997년도 추계학술대회 논문집
• /
• pp.70-75
• /
• 1997

This paper proposes a new approach to the design of fuzzy-neuro control for track vehicle system using fuzzy logic based on neural network. The proposed control scheme uses a Gaussian function as a unit function in the neural network-fuzzy, and back propagation algorithm to train the fuzzy-neural network controller in the framework of the specialized learning architecture. It is proposed a learning controller consisting of two neural network-fuzzy based of independent reasoning and a connection net with fixed weights to simply the neural networks-fuzzy. The performance of the proposed controller is illustrated by simulation for trajectory tracking of track vehicle speed.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1997년도 춘계학술대회 논문집
• /
• pp.133-139
• /
• 1997

This paper presents a new approach to the dynamic control technique for track vehicle system using neural network-fuzzy control method. The proposed control scheme uses a Gaussian function as a unit function in the neural network-fuzzy, and back propagation algorithm to train the fuzzy-neural network controller in the framework of the specialized learning architecture. It is propored a learning controller consisting of two neural network-fuzzy based on independent resoning and a connection net with fixed weights to simply the neural network-fuzzy. The performance of the proposed controller is shown by simulation for trajectory tracking of the speed and azimuth of a track vehicle

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 1999년도 추계학술대회 논문집 - 한국공작기계학회
• /
• pp.142-147
• /
• 1999

This paper presents a new approach to the dynamic control technique for track vehicle system using neural network-fuzzy control method. The proposed control scheme uses a Gaussian function as a unit function in the neural network-fuzzy, and back propagation algorithm to train the fuzzy-neural network controller in the framework of the specialized learning architecture. It is proposed a learning controller consisting of two neural network-fuzzy based on independent reasoning and a connection net with fixed weights to simply the neural networks-fuzzy. The performance of the proposed controller is shown by simulation for trajectory tracking of the speed and azimuth of a track vehicle.

• 로봇학회논문지
• /
• 제17권2호
• /
• pp.133-141
• /
• 2022

This paper introduces model compression algorithms which make a deep neural network smaller and faster for embedded systems. The model compression algorithms can be largely categorized into pruning, quantization and knowledge distillation. In this study, gradual pruning, quantization aware training, and knowledge distillation which learns the activation boundary in the hidden layer of the teacher neural network are integrated. As a large deep neural network is compressed and accelerated by these algorithms, embedded computing boards can run the deep neural network much faster with less memory usage while preserving the reasonable accuracy. To evaluate the performance of the compressed neural networks, we evaluate the size, latency and accuracy of the deep neural network, DenseNet201, for image classification with CIFAR-10 dataset on the NVIDIA Jetson Xavier.

• 한국정밀공학회지
• /
• 제13권4호
• /
• pp.75-86
• /
• 1996

In this paper, we propose a new fuzzy-neural network control scheme for the speed and azimuth control of a mobile robot. The proposed control scheme uses a gaussian function as a unit function in the fuzzy-neural network, and back propagation algorithm to train the fuzzy-neural network controller in the frame-work of the specialized learning architecture. It is proposed a learning controller consisting of two fuzzy-neural networks based on independent reasoning and a connection net woth fixed weights to simply the fuzzy-neural network. The effectiveness of the proposed controller is illustrated by performing the computer simulation for a circular trajectory tracking of a mobile robot driven by two independent wheels.

• 대한의용생체공학회:의공학회지
• /
• 제42권4호
• /
• pp.167-174
• /
• 2021

The purpose of this study is to classify TIFF images, PNG images, and JPEG images using deep learning, and to compare the accuracy by verifying the classification performance. The TIFF, PNG, and JPEG images converted from chest X-ray DICOM images were applied to five deep neural network models performed in image recognition and classification to compare classification performance. The data consisted of a total of 4,000 X-ray images, which were converted from DICOM images into 16-bit TIFF images and 8-bit PNG and JPEG images. The learning models are CNN models - VGG16, ResNet50, InceptionV3, DenseNet121, and EfficientNetB0. The accuracy of the five convolutional neural network models of TIFF images is 99.86%, 99.86%, 99.99%, 100%, and 99.89%. The accuracy of PNG images is 99.88%, 100%, 99.97%, 99.87%, and 100%. The accuracy of JPEG images is 100%, 100%, 99.96%, 99.89%, and 100%. Validation of classification performance using test data showed 100% in accuracy, precision, recall and F1 score. Our classification results show that when DICOM images are converted to TIFF, PNG, and JPEG images and learned through preprocessing, the learning works well in all formats. In medical imaging research using deep learning, the classification performance is not affected by converting DICOM images into any format.

• Journal of Biosystems Engineering
• /
• 제16권1호
• /
• pp.90-99
• /
• 1991

• 한국측량학회지
• /
• 제39권3호
• /
• pp.141-148
• /
• 2021

본 논문은 SegNet과 ResNet을 조합한 딥러닝을 이용하여 횡단보도를 검출하는 방법을 제안한다. 시각 장애인의 경우 횡단보도가 어디에 있는지 정확히 아는 게 안전한 교통 시스템에서 중요하다. 딥러닝에 의한 횡단보도 검출은 이 문제에 대한 좋은 해결책이 될 수 있다. 로봇 시각 기반 보조 기술은 지난 몇년 동안 카메라를 사용하는 특정 장면에 초점을 두고 제안되어 왔다. 이러한 전통적인 방법은 비교적 긴 처리 시간으로 의미있는 결과를 얻었으며 횡단보도 인식을 크게 향상시켰다. 그러나 전통적인 방법은 지연 시간이 길고 웨어러블 장비에서 실시간을 만족시킬 수 없다. 본 연구에서 제안하는 방법은 취득한 영상에서 횡단보도를 빠르고 안정적으로 검출하기 위한 모델을 제안한다. 모델은 SegNet과 ResNet을 기반으로 개선되었으며 3단계로 구성된다. 첫째, 입력 영상을 서브샘플링하여 이미지 특징을 추출하고 ResNet의 컨벌루션 신경망을 수정하여 새로운 인코더로 만든다. 둘째, 디코딩 과정에서 업샘플링 네트워크를 통해 특징맵을 원영상 크기로 복원한다. 셋째, 모든 픽셀을 분류하고 각 픽셀의 정확도를 계산한다. 이 실험의 결과를 통하여 수정된 시맨틱 분할 알고리즘의 적격한 정확성을 검증하는 동시에 결과 출력 속도가 비교적 빠른 것으로 파악되었다.

• 대한전자공학회논문지
• /
• 제27권6호
• /
• pp.972-981
• /
• 1990

Many researchers have been interested in whether complex computational problems can be solved by the neural net or not. Especially, problems of early vision are integrated by Tikhonov's regularization theory. Regularization technique can be realized in resistive net. In this paper, we suggest the resistive net with upper and lower thresholder to be able to compute shape from shading and to solve its discontinuous problem. We simulate three algorithms-Horn's algorithm, resistive net and up-low thrwsholding net -with respect to three cases-fully boundary, boundary losing partly and noisy image. As being able to cope with crease and discontinuous parts, we get the good 3D shape from shading.