Computer vision, robot perception, and multi-sensor fusion.
- AgriFoRwArdS CDT Annual Conference (2022): Automatic aphid counting based on yellow water pan trap imagery and deep learning.
Activities and Outputs
- Represented the CDT at the Douglas Bomford Trust bi-annual meeting (Mar 2022)
- Represented the CDT at the school outreach University of Lincoln British Science Week event 2022.
My name is Xumin Gao, I come from China. I have studied and did research work at the Institute of Robotics and Intelligent Systems, Wuhan University of Science and Technology. My research was mainly focused on computer vision and robot perception.
After graduating from university, I worked in an autonomous vehicle company, and worked on the algorithm development of computer vision, including vehicle recognition, vehicle feature point extraction. Then I worked in an intelligent agricultural technology company. I was mainly responsible for image-based poisonous weed detection and segmentation for autonomous weeding robots and UAVs, as well as satellite imagery segmentation for farmland monitoring.
I am a robot lover. At present, I have made many robots, including dancing robots, indoor service robots, weeding robots and so on. If you want to see these lovely robots, you can visit this website. In my spare time, I especially like dancing (I can dance at least five different dances), hiking and exploring some natural life. Sometimes, I also like to record my life and feelings by writing.
The reason why I chose to join the AgriFoRwArdS CDT is that this project is especially close to my research area of interest. In addition, the work experience I have makes me realise that intelligent agricultural robots have great potential development space at present. I will be studying my PhD at the University of Lincoln. I look forward to meeting other AgriFoRwArdS CDT members and working with them. At the same time, I believe I will have a good time in Lincoln.
Automatic aphid counting based on yellow water pan trap imagery and deep learning
Aphids can cause direct damage and indirect virus transmission to crops. Timely monitoring the number of aphids can prevent the large-scale outbreak of aphids. However, the manual counting of aphids which is commonly used at present is inefficient, and it requires professional staff to complete it. Therefore, this project designs an automatic aphid counting network based on deep learning to replace the manual counting. Moreover, on this basis, we will challenge the common difficulties of automatic aphids counting, including dense distribution of aphids, tiny size of aphids and so on. It can monitor aphids automatically and accurately, so as to protect crop growth and improve crop yields.
Title to be confirmed