报告时间：北京时间2021年09月22日（星期三）: 15:00-18:00

China-Montenegro "Flexible sensing and compressed sensing signal processing technology" International Online Seminar

Time:

15:00 (GMT+8), Sep 22, 2021, Wednesday, Beijing Time

9:00 am (GMT+2) Sep 22, 2021, Wednesday, Montenegro Time

Online Meeting: Mircosoft Teams

https://teams.microsoft.com/l/meetup-join/19:meeting_YTNhZjgyYTUtZDIxMS00YzI1LTg0YzItODZhNjFkODY1ZmNj@thread.v2/0?context=%7B%22Tid%22:%22591e5439-291e-418a-9759-245c6ce4e6ed%22,%22Oid%22:%22be1894d9-be36-4b3d-a7e4-2d9f69914bce%22%7D

Chinese participants:

1. Title:           Optimization of texture profile analysis compression ratios for chilled mutton based on significant correlation analysis 

Author:            ZHANG Guoxian, Post Ph.D researcher, Beijing Laboratory of Food Quality and Safety, College of Information and Electrical Engineering, China Agricultural University, China

Short Abstract:

Texture profile analysis (TPA) is widely used in food quality detection. Appropriate detection conditions are the key factor to obtain accurate and effective TPA index data. A framework was proposed, through which the texture information of chilled mutton could be reliably obtained by optimizing test conditions of TPA. In the presentation, a relevant significance marker method was performed to objectively elaborate the changing law of TPA indexes data under different compression ratios. The optional compression ratio was determined by a multi-level optimization model for TPA indexes data. Finally, the results were verified by using the correlation between TPA indexes and physiological indexes of chilled mutton..

2. Title:          Wearable technology and animal health monitoring 

Author:           ZHANG Mengjie, Ph.D candicate, Beijing Laboratory of Food Quality and Safety, College of Engineering, China Agricultural University, China

Short Abstract:   

At present, the research on animal physiological index detection mostly focuses on static biochemical detection, which can not achieve real-time and continuous recording and transmission of information.Many sensors used for animal health management are at different stages of commercialization. Some technologies that can produce accurate health and disease diagnosis are only applicable to humans, and few modifications or tests are performed on animal models.Bioinformatics sensors can quantify a variety of physiological and behavioral responses of livestock and poultry, and provide significant benefits and applications in health monitoring, reproduction, and feeding.Wearable Internet of things technology integrates sensor, wireless communication, data management platform, etc., and is expected to become the most effective and feasible technical scheme for animal health supervision.

3. Title:           Live Oyster detection, life prediction and evaluation method based on inkjet printing flexible sensing technology

Author:           LIU Pengfei , Ph.D, Beijing Laboratory of Food Quality and Safety, College of Engineering, China Agricultural University, China

Short Abstract:  

The flexible sensor based on inkjet printing has the advantages of high sensitivity, low temperature impact, high spatial resolution and easy signal detection. Flexible sensors have the advantages of fast response, high accuracy and good reliability in pressure, temperature and humidity. Through the research and debugging of the resistance type flexible pressure sensor based on copper nanowires/graphene/melamine foam, it has important theoretical and practical significance for the oyster activity detection and prediction evaluation through the test pressure.

4. Title:            Portable flexible sensor system for bioimpedance signal detection

Author:           ZHANG Luwei,   Ph.D candicate, Beijing Laboratory of Food Quality and Safety, College of Engineering, China Agricultural University, China

Short Abstract:  

Bioimpedance is a noninvasive technique that denotes the passive electrical properties of biological material. When live organisms undergo a series of physiological changes such as maturation, senescence, and cellular dysfunction, biological tissue exposed under alternating current exhibits complex impedance behavior for the detection of rapid changes associated with the positive response of the biological organism. In this report, a portable flexible sensor system for bioimpedance signal detection is presented with a view to achieving rapid evaluation of changes in the physiological quality of live organisms.

Montenegro participants

1. Title:      Noise removal based on compressive sensing principles 

Author:            Miloš Brajović, Ph.D, Faculty of Electrical Engineering, University of Montenegro.  

Short Abstract:

Signal samples or image pixels can be so heavily corrupted by disturbances, that it is reasonable to consider these values as unavailable. If a signal or an image is sparse in a transformation domain, we can accurately reconstruct these excluded values. We will present two approaches for automatic detection and reconstruction of disturbed samples (pixels). One is based on a gradient of concentration measure, whereas the other is inspired and uses the RANSAC algorithm. These approaches successfully deal with challenging situations when the disturbance is within the signal values, or within the range of pixel values. The approaches work blindly, without presumptions regarding the statistical characteristics or intensity of the noise.

2. Title:          Quick Response Code Recovery using Sparse Signal Processing Approach 

Author:            MAJA LAKIČEVIĆ ŽARIĆ, Ph.D, Faculty of Electrical Engineering, University of Montenegro.  

Short Abstract:   

In this research, coding performance when the code image is insufficiently sampled is examined. The QR code was used as an example, as its popularity is increasing every day. The procedure of random insufficient sampling is motivated by the compression sensing scenario. The obtained results prove that the code encoding can be successfully done after reconstruction even when 70% of the pixels are missing.

3. Title:          Compressive Sensing Approach in Multimedia, Biomedicine and Communications

Author:           Anđela Draganić, Ph.D, Faculty of Electrical Engineering, University of Montenegro.

Short Abstract:  

Compressive sensing is a new theory and concept that opens the possibility to acquire a significantly smaller amount of data, but to be able to achieve the same quality of the final information as it is the case when the entire physical phenomenon is sensed. Therefore, the research efforts are made to simplify the very expensive devices and apparatus for data recording, imaging, sensing (for instance MRI scanners, PET scanners for Computed tomography, high resolution cameras, etc.). It is also important to reduce drastically the data acquisition time, even up to almost 20% of the currently required time.

Contact: Jinchao XU;  xuvic@cau.edu.cn