Acta Optica Sinica, Volume. 39, Issue 8, 0810002(2019)
Substation Infrared Image Segmentation Based on Novel Threshold Selection Method
Fig. 3. Segmentation method. (a) Composite image; (b) our method; (c) traditional; (d) filter smoothing; (e) threshold selection process
Fig. 4. Comparison of results of different segmentation methods. (a) Fuzzy entropy segmentation; (b) maximum type variance method; (c) Otsu; (d) K-means; (e) fuzzy threshold segmentation; (f) level set segmentation
Fig. 5. Current transformer fault. (a) Infrared image; (b) our method segmentation; (c) fuzzy threshold segmentation; (d) K-means segmentation; (e) fuzzy entropy segmentation
Fig. 6. Cable fault. (a) Infrared image; (b) our method segmentation; (c) fuzzy threshold segmentation; (d) K-means segmentation; (e) fuzzy entropy segmentation
Fig. 7. Corrosion comparison. (a) Uncorroded current transformer; (b) traditional corrosion of current transformer; (c) our corrosion of current transformer; (d) uncorroded cable; (e) traditional corrosion of cable; (f) our corrosion of cable
|
Get Citation
Copy Citation Text
Qingsheng Zhao, Yuying Wang, Xuping Wang, Zun Guo. Substation Infrared Image Segmentation Based on Novel Threshold Selection Method[J]. Acta Optica Sinica, 2019, 39(8): 0810002
Category: Image Processing
Received: Jan. 15, 2019
Accepted: Mar. 21, 2019
Published Online: Aug. 7, 2019
The Author Email: Qingsheng Zhao (zhaoqs1996@163.com)