Laser & Optoelectronics Progress, Volume. 59, Issue 2, 0210016(2022)
Optimization Method for Infrared Eye Movement Image Segmentation
Fig. 1. Schematic diagram of the label of the eye area image. (a) Original image; (b) label image
Fig. 2. Structure of the S-RITnet
Fig. 3. Structure of the ESRGAN
Fig. 4. Reproduction results of eye area images of different algorithms. (a) Original image; (b) bilinear interpolation; (c) bicubic interpolation; (d) ESRGAN
Fig. 5. Comparison of image equalization effect. (a) Orignial image; (b) linear transformation; (c) Gamma transformation; (d) HE; (e) CLAHE
Fig. 6. Structure of the RITnet
Fig. 7. Experiment evaluation curve and segmentation results. (a) MIoU curve of the validation set; (b) segmentation results of each data set
Fig. 8. Segmentation results of the RITnet data set. (a) Original image; (b) BE50-RITnet; (c) BE100-RITnet; (d) S-RITnet (NE); (e) label
|
|
|
Get Citation
Copy Citation Text
Chang Li, Yu Liu, Jinglin Sun. Optimization Method for Infrared Eye Movement Image Segmentation[J]. Laser & Optoelectronics Progress, 2022, 59(2): 0210016
Category: Image Processing
Received: Dec. 2, 2020
Accepted: Mar. 16, 2021
Published Online: Dec. 23, 2021
The Author Email: Liu Yu (liuyu@tju.edu.cn)