Faculty Profiles - ZHANG Lifeng

Papers - ZHANG Lifeng

Division display61 - 80 of about 223／ All the affair displays >>

Proposal of Information Addition Method to Existing Signs for Autonomous Driving Reviewed

Koho Futatsuishi, Lifeng Zhang

Proceedings of the 9th IIAE International Conference on Industrial Applications Engineers 2021 ( The Institute of Industrial Applications Engineers ) 1 ( 1 ) 101 - 106 2021.03

Language：English Publishing type：Research paper (international conference proceedings)

Japan Kitakyushu 2021.03.26 - 2020.03.30

Automatic driving technology is developing in recent years. High level automatic driving use raders and cameras,etc to grasep the surrounding environment. Recognition of road signs is one of them. However, in the recognition of the sign, there is a problem that a similar shape things is erroneously recognized in the current method of judging by an external feature. In this research, we propose a system that recognizes road signs by reading a QR code made of a polarizing plate.

DOI： 10.12792/iciae2021.019

Other Link： https://www2.ia-engineers.org/conference/index.php/iciae/iciae2021/paper/view/2405

Prevention of Forgetting Shopping Items in Response to Shopping List Ambiguity Reviewed

Yoshihisa Igimi, Lifeng Zhang

Proceedings of the 9th IIAE International Conference on Industrial Applications Engineers 2021 ( The Institute of Industrial Applications Engineers ) 1 ( 1 ) 97 - 100 2021.03

Language：English Publishing type：Research paper (international conference proceedings)

Japan Kitakyushu 2021.03.26 - 2020.03.30

Shopping plays a critical role in daily life, and also it is a time-consuming activity. For the coming elder society in Japan, the elders who live alone have to go shopping by themselves and often forget to buy the target goods is in their shopping list. This forgetting is because the shopping list is often made with ambiguous expressions compared to the store registered product name. This research proposes a new shopping item forgetting prevention system that can detect the not buying item immediately after purchasing to dealing with the shopping list's ambiguity.

DOI： 10.12792/iciae2021.018

Other Link： https://www2.ia-engineers.org/conference/index.php/iciae/iciae2021/paper/view/2401

Stereo Matching Based on Features of Image Patch Reviewed

Shi Zhou, He Li, Miaomiao Zhu, Zhen Li, Mitsunori Mizumachi, Lifeng Zhang

Proceedings of the 9th IIAE International Conference on Industrial Applications Engineers 2021 ( The Institute of Industrial Applications Engineers ) 1 ( 1 ) 18 - 22 2021.03

Language：English Publishing type：Research paper (international conference proceedings)

Japan Kitakyushu 2021.03.26 - 2020.03.30

Stereo matching is a branch of 3D vision and has a wide range of applications in 3D reconstruction and autonomous driving. Recently, stereo matching methods leverage the information of two full image to calculate disparity map. However, these methods still have difficulties in texture-less regions and occlusion regions, and post-processing is used to improve accuracy. Therefore, there is a large computational cost in the feature extraction and post-processing. In this paper, we propose a stereo matching method based only on features of image patches and predict the disparity of region without occlusion. And post-processing is performed to modify all kind of mismatching based on the correct disparity. Furthermore, we evaluated our proposed method on the Middlebury dataset. The results show that our method performs well in all areas.

DOI： 10.12792/iciae2021.006

Other Link： https://www2.ia-engineers.org/conference/index.php/iciae/iciae2021/paper/view/2423

Proposal of a System for the Visually Impaired to Identify Store Front Merchandise with no Assistance Reviewed

Tomoyuki Zaizen, Lifeng Zhang

Proceedings of the 9th IIAE International Conference on Industrial Applications Engineers 2021 ( The Institute of Industrial Applications Engineers ) 1 ( 1 ) 13 - 17 2021.03

Language：English Publishing type：Research paper (international conference proceedings)

Japan Kitakyushu 2021.03.26 - 2020.03.30

In recent years, the aging of the population in Japan has become a problem. In addition, many elderly people are visually impaired, and the number of visually impaired people is expected to increase in the future. However, the lack of caregivers is also a problem in an aging society, and the problem of not having enough caregivers for the increasing number of visually impaired people is expected to arise in Japan in the future. Due to the shortage of assistants, elderly visually impaired people tend to avoid going out, and their safety confirmation may be delayed. That also becomes a big problem. Therefore, in order to prevent them, it is necessary to have a system that enables the visually impaired to live without nursing care. In this study, we propose a system that enables the visually impaired to buy products without nursing care, which is based on an image processing technique and machine learning.

DOI： 10.12792/iciae2021.005

Other Link： https://www2.ia-engineers.org/conference/index.php/iciae/iciae2021/paper/view/2399

A Proposal for Creating Syllabic Datasets for Japanese Language Lipreading by Using Machine Learning Reviewed

Rui Kitahara, Lifeng Zhang

Proceedings of the 9th IIAE International Conference on Industrial Applications Engineers 2021 ( The Institute of Industrial Applications Engineers ) 1 ( 1 ) 8 - 12 2021.03

Language：English Publishing type：Research paper (international conference proceedings)

Japan Kitakyushu 2021.03.26 - 2020.03.30

Although lip-reading using image processing and machine learning has been mainly performed at the word level, it has been shown that using LipNet, a network that enables recognition at the sentence level, improves the recognition accuracy over the former method. However, this was the case for English speakers. In this study, the data set was created based on speech scenes containing all 50 Japanese sounds, and the recognition accuracy was evaluated using LipNet.

DOI： 10.12792/iciae2021.004

Other Link： https://www2.ia-engineers.org/conference/index.php/iciae/iciae2021/paper/view/2397

3D Object Detection Algorithms on Intelligent Driving: A Survey Reviewed

Zhen Li, Miaomiao Zhu, Shi Zhou, Lifeng Zhang

PROCEEDINGS OF THE TWENTY-SIXTH INTERNATIONAL SYMPOSIUM ON ARTIFICIAL LIFE AND ROBOTICS (AROB 26th 2021) ( International Society of Artificial Life and Robotics ) 1 ( 1 ) 38 - 43 2021.01

Language：English Publishing type：Research paper (international conference proceedings)

Japan Online 2021.01.21 - 2021.01.23

With the rapid development of Artificial Intelligent algorithms on Computer Vision, 2D object detection has greatly succeeded and been applied in various industrial products. In the past several years, the accuracy of 2D object detection has been dramatically improved, even beyond the human eyes' detection ability. However, there is still a limitation of 2D object
detection using for the application of Intelligent Driving. A safe and reliable self-driving car needs to detect a 3D model of the around objects so that an intelligent driving car has a perception ability to real driving situations. This paper systematically surveys the development of 3D object detection methods applied to intelligent driving technology. This paper also analyzes the shortcomings of the existing 3D detection algorithms and the future development direction of 3D detection algorithms on intelligent driving.

Self-supervised Monocular Depth Estimation with Occlusion Mask and Edge Awareness Reviewed

Shi Zhou, Miaomiao Zhu, Zhen Li, Mitsunori Mizumachi, Lifeng Zhang

PROCEEDINGS OF THE TWENTY-SIXTH INTERNATIONAL SYMPOSIUM ON ARTIFICIAL LIFE AND ROBOTICS (AROB 26th 2021) ( International Society of Artificial Life and Robotics ) 1 ( 1 ) 404 - 408 2021.01

Language：English Publishing type：Research paper (international conference proceedings)

Japan Online 2021.01.21 - 2021.01.23

Depth estimation is one of the basic and important tasks in 3D vision. Recently, many works have been done in selfsupervised depth estimation based on geometric consistency between frames. However, these research works still have difficulties in ill-posed regions, such as occlusion areas and texture-less areas. This work proposes a novel self-supervised monocular depth estimation method based on occlusion mask and edge awareness to overcome these difficulties. The occlusion mask divides the image into two classes, making the training of the network more reasonable. The edge awareness loss function is designed based on the edge obtained by the traditional method, so that the method has strong robustness to various lighting conditions. Furthermore, we evaluated our proposed method on the KITTI dataset. The occlusion mask and edge awareness are both beneficial to find corresponding points in ill-posed regions.detection using for the application of Intelligent Driving. A safe and reliable self-driving car needs to detect a 3D model of the around objects so that an intelligent driving car has a perception ability to real driving situations. This paper systematically surveys the development of 3D object detection methods applied to intelligent driving technology. This paper also analyzes the shortcomings of the existing 3D detection algorithms and the future development direction of 3D detection algorithms on intelligent driving.

Person Re-identification in the Real-World Application Based on Deep Learning Reviewed

Miaomiao Zhu, Shengrong Gong, Zhenjiang Qian, Seiichi Serikawa, and Lifeng Zhang

PROCEEDINGS OF THE TWENTY-SIXTH INTERNATIONAL SYMPOSIUM ON ARTIFICIAL LIFE AND ROBOTICS (AROB 26th 2021) ( International Society of Artificial Life and Robotics ) 1 ( 1 ) 228 - 233 2021.01

Language：English Publishing type：Research paper (international conference proceedings)

Japan Online 2021.01.21 - 2021.01.23

Person re-identification (ReID) aims at retrieving a person of interest across multiple non-overlapping cameras. Because of increasing demand for real-world applications in intelligent video surveillance, person re-identification has become an important task in computer vision, and achieved high performance in recent years. However, the traditional research is mainly
focus on matching cropped pedestrian images between queries and candidates on commonly used datasets, and divided into two steps: pedestrian detection and person re-identification, there is still a big gap with practical applications. Under the premise of model optimization, based on the existing object detection and person re-identification, this paper achieves a one-step search of specific pedestrians in whole images or video sequences in actual application scenario. The experimental results show that our method is effective in commonly used datasets, and has achieved good results in the real-world application, such as finding criminals, cross-camera person tracking and activity analysis.

Proposal of Japanese Syllabary Data Set Construction for Japanese Lip Reading using Machine Learning International journal

Kitahara Rui, Zhang Lifeng

Proceedings of IIAE Annual Conference ( The Institute of Industrial Applications Engineers ) 2021 ( 0 ) 28 - 29 2021.01

Development of Virtual Marker Type Automatic Tracking Projection Mapping System by Image Processing Reviewed

Masaki Inoue, Lifeng Zhang

Proceedings of International Conference on Technology and Social Science 2020 (ICTSS 2020) 2020.12

Language：English Publishing type：Research paper (international conference proceedings)

日本 Kiryu, Japan 2020.12.02 - 2020.12.04

In recent years, automatic tracking projection mapping that maps moving objects such as people and balls has also become extensive.
As a principle of auto-following projection mapping, using a marker for specifying the position of a projection target is common. The position is estimated by reflecting the infrared rays emitted from the projector by the marker attached to the projection target.
This research has developed a tracking projection mapping system that does not require physical markers but can realize a more flexible image expression.
The system performs follow-up mapping without physical markers by virtualizing markers that estimate the projection target's position, a three-dimensional figure such as a cube or rectangular parallelepiped, by image processing. Specifically, we have realized a target position estimation system that does not require physical markers by accurately detecting the four corners of the projection target by performing corner detection step by step using ROI.

Research on Image Privacy Protection Algorithm Based on Generative Adversarial Network Reviewed International journal

Wu Zhiqiang, Zhang Lejun, Zhang Lifeng, Seiichi Serikawa

2020 2nd International Conference on Machine Learning, Big Data and Business Intelligence (MLBDBI) ( IEEE ) 500 - 504 2020.10

Language：English Publishing type：Research paper (international conference proceedings)

China Taiyuan 2020.10.23 - 2020.10.25

In view of the existing privacy protection means, the sensitive areas in the pictures, such as face, text and other parts, are changed to protect privacy, or the objects involving privacy are removed from the picture or video. However, the obvious defect of these methods is that they destroy the practicability of the photos. When the privacy information in the photos is removed, the important information contained in the photos is also removed. In order to solve this problem, this paper introduces the generative adversarial network and proposes an image privacy protection algorithm. This algorithm defines information privacy methods and optimizes limited areas to ensure that information is properly protected. Different image protection schemes are designed to adapt and not adapt to the safe path. In order to meet the needs of privacy protection, we apply the image privacy protection algorithm in the generative adversarial network. Finally, the simulation analysis of the privacy protection algorithm based on the generative adversarial network is carried out. The experimental results show that compared with the encryption algorithm, the privacy protection algorithm based on the generative adversarial network has better performance in the end-to-end delay and encryption speed.

DOI： 10.1109/MLBDBI51377.2020.00105

Other Link： https://ieeexplore.ieee.org/document/9361051

Person Re-identification on Mobile Devices Based on Deep Learning Reviewed

Miaomiao Zhu, Shengrong Gong, Zhenjiang Qian, Lifeng Zhang

Proceedings of the 8th IIAE International Conference on Industrial Applications Engineers 2020 ( The Institute of Industrial Applications Engineers ) 1 ( 1 ) 253 - 260 2020.03

Language：English Publishing type：Research paper (international conference proceedings)

Japan Matsue 2020.03.26 - 2020.03.30

Person Re-identification as an important supplement to face recognition, refers to a network of cameras with non-overlapping vision domains, through the use of computer vision technology to solve the cross-camera and cross-scene pedestrian recognition and retrieval. That is, determine whether there is a specific pedestrian to be detected in the different images or different video sequences. At present, many person re-identification research works mainly carried out through experimental verification and evaluation on large ReID datasets such as Market-1501, DukeMTMC-reID, MSMT17, and CUHK03. In this paper, based on the existing deep-learning person re-identification research, combines with the actual application scenarios, under the premise of analyzing the technical feasibility, we propose a complete process for Person Re-identification based on mobile devices, aims to combine pedestrian detection and person re-identification to perform real-time pedestrian detection and query. In this process, not only the features extracted by pedestrians can be reused, but also the research on person re-identification can be better and effectively applied, such as tracking criminals and searching for missing children.

DOI： 10.12792/iciae2020.044

Other Link： https://www2.ia-engineers.org/conference/index.php/iciae/iciae2020/paper/view/2351

AI security--The Application of Adversarial Samples in Face Recognition Reviewed

ZhangLe Jun, Wu Zhiqiang, Zhang Lifeng, Seiichi Serikawa

Proceedings of the 8th IIAE International Conference on Industrial Applications Engineers 2020 ( The Institute of Industrial Applications Engineers ) 1 ( 1 ) 336 - 340 2020.03

Language：English Publishing type：Research paper (international conference proceedings)

Japan Matsue 2020.03.26 - 2020.03.30

With the development of economy and the advancement of science and technology, society's requirements for the accuracy of face recognition have gradually increased, and its dependence on face recognition technology has become stronger and stronger. When deploying a neural network in a high-security environment, it is usually easier to ignore the security threat caused by vulnerability. Fully analyze the vulnerability of traditional face recognition technology, and use the generated adversarial samples to design a novel eyeglass patch sample, which can successfully deceive the face recognition system based on convolutional neural network. In addition, the security of artificial intelligence can bring new opportunities for face recognition. Applying artificial intelligence technology to face recognition can effectively improve accuracy. The research purpose of this paper is to analyze the application of adversarial samples in face recognition based on the security of artificial intelligence in order to solve the problems of low security and high vulnerability in traditional face recognition.

DOI： 10.12792/iciae2020.057

Other Link： https://www2.ia-engineers.org/conference/index.php/iciae/iciae2020/paper/view/2329

Face Recognition with Age, Gender and Emotion Estimations Reviewed

Dongqin Gu, Miaomiao Zhu, Lifeng Zhang, Seiichi Serikawa

Proceedings of the 8th IIAE International Conference on Industrial Applications Engineers 2020 ( The Institute of Industrial Applications Engineers ) 1 ( 1 ) 324 - 329 2020.03

Language：English Publishing type：Research paper (international conference proceedings)

Japan Matsue 2020.03.26 - 2020.03.30

Based on the residual network, this paper proposes a system for face recognition with gender, age and emotion estimations. It can be implemented in different retail industries, helping merchants build their own databases, analyze consumers' shopping preferences with different ages and genders, track users' sentiment, feedback users' experience, and adjust purchase lists and operating modes in a timely manner.

DOI： 10.12792/iciae2020.055

Other Link： https://www2.ia-engineers.org/conference/index.php/iciae/iciae2020/paper/view/2321

Optimization for greedy non-maximum suppression based on multi-task convolutional neural network Reviewed

Qingqing Hong, Qian Fan, Lifeng Zhang

Journal of the Institute of Industrial Applications Engineers ( The Institute of Industrial Applications Engineers ) 8 ( 1 ) 39 - 45 2020.01

Language：English Publishing type：Research paper (scientific journal)

Non-maximum suppression (NMS) is an essential part of the face detection pipeline based on a convolutional neural network (CNN). The typical approach for NMS used by face detection is a greedy, locally optimal strategy, which is to localize objects from a set of candidate locations. However, NMS still has some shortcomings, such as sometimes the detection box has no relationship with a high classification score, which leads to misjudging face localization during NMS. In this paper, we observed that the NMS implemented on the multi-task convolutional neural network (MTCNN) is a cascaded network and the enhanced NMS based on MTCNN to achieve high performance during face detection and alignment. We employ WIDER FACE as the test dataset to evaluate our proposal. The precision and recall curves are drawn when using three subsets at different thresholds. The result shows that the proposed approach can perform better performance than traditional NMS.

DOI： 10.12792/JIIAE.8.39

Other Link： https://www2.ia-engineers.org/Journal_E/index.php/jiiae/article/view/197

An Application of Multi-Face Recognition Based on Feature Extraction Reviewed

Dongqin Gu, Seiichi Serikawa, Lifeng Zhang

Proceedings of the 7th IIAE International Conference on Intelligent Systems and Image Processings 2019 ( The institute of Industrial Applications Engineers ) 207 - 211 2019.09

Research for Improving Identification Accuracy of Specific Fish Species with CNN Reviewed

Yuta Shoji, Lifeng Zhang

Proceedings of the 7th IIAE International Conference on Intelligent Systems and Image Processings 2019 ( The institute of Industrial Applications Engineers ) 190 - 193 2019.09

Preprocessing Method for Training Dataset in Character Recognition using Convolutional Neural Networkbase Reviewed

Akase Taku, Hitoshi Nakao, Lifeng Zhang

Proceedings of the 7th IIAE International Conference on Intelligent Systems and Image Processings 2019 ( The institute of Industrial Applications Engineers ) 186 - 189 2019.09

Proposal of Character Correction Method using Database Reviewed

Takahiro Fujishima, Ken Kariya, Lifeng Zhang

Proceedings of the 7th IIAE International Conference on Intelligent Systems and Image Processings 2019 ( The institute of Industrial Applications Engineers ) 178 - 181 2019.09

Development of OCR Mobile Application Including Miss-recognized Proofreading System Using Database Search Algorithm Reviewed

Ken Kariya, Takahiro Fujishima, Lifeng Zhang

Proceedings of the 7th IIAE International Conference on Intelligent Systems and Image Processings 2019 ( The institute of Industrial Applications Engineers ) 182 - 185 2019.09

Authorship：Corresponding author Language：English Publishing type：Research paper (international conference proceedings)

日本北九州市 2019.09.05 - 2019.09.09

DOI： 10.12792/icisip2019.034

Other Link： https://www2.ia-engineers.org/conference/index.php/icisip/icisip2019/paper/view/2194/1419

　 PREV - NEXT 　

<ZHANG Lifeng>

Papers - ZHANG Lifeng