Optimization on multi-object tracking and segmentation in pigs’ weight measurement
| Autor: | Ximeng Li, Yulong Qiao, Chunyu Chen, Hengxiang He, Zhang Xingfu |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
0106 biological sciences
Computer science business.industry Forestry 04 agricultural and veterinary sciences Horticulture Tracking (particle physics) 01 natural sciences Computer Science Applications Convolution Feature (computer vision) Video tracking 040103 agronomy & agriculture 0401 agriculture forestry and fisheries Embedding Segmentation Computer vision Deconvolution Artificial intelligence business Agronomy and Crop Science Spatial analysis 010606 plant biology & botany |
| Zdroj: | Computers and Electronics in Agriculture. 186:106190 |
| ISSN: | 0168-1699 |
| Popis: | Weight of pigs is highly correlated to their health. At present, 3D cameras can get spatial information, which develop non-contacting weight measurement. Separating pigs from the background is the first step, and tracking in a short video can make the weight more accurate than predicting weight on single image. Multi-Object Tracking and Segmentation (MOTS) in a video has received more attention with adding association embedding branch into instance segmentation network. Despite its success, the MOTS network has a crucial problem in practical application, that the predicted masks do not fit the objects well. The reason is low resolution of the feature maps in mask branch. So we improve the mask generation branch by cascading deconvolution layer and atrous convolution layer flexibly. The experimental results show that two deconvolution layers cooperating with two atrous convolution layers perform better. In pigs’ weight measurement, this method outputs more precise masks than original network. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect