RGB color decomposition and image feature extraction of flame image in rear of sintering machine
| Autor: | Wu Chen, Hou Qingwen, Wang Fubin, Zhang Meng, Chen Zhikun, Yang Haolong, Chen Xianzhong |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Materials science
business.industry Binary image 010401 analytical chemistry Feature extraction Top-hat transform 02 engineering and technology Image segmentation 01 natural sciences 0104 chemical sciences Image texture Feature (computer vision) 0202 electrical engineering electronic engineering information engineering RGB color model 020201 artificial intelligence & image processing Computer vision Artificial intelligence business Feature detection (computer vision) |
| Zdroj: | 2017 36th Chinese Control Conference (CCC). |
| Popis: | The stability of the sintering end point is an important precondition for the smooth running of the sintering machine and the assurance of the quality and quantity of sintering ore. The sintering flame of the sintering machine tail involves a great deal of associated information about sintering end point. Extracting the image feature of the flame has important significance for judging the operation state of sintering. Firstly, taking flame image of normal sintering as research objects, analyzing the energy information in the flame image; Secondly, reducing the halo effect on the image by processing image with Discrete Fourier transform (DFT); Thirdly, using the color decomposition method to decompose the RGB image into three components of R, G and B, obtaining the principal component image of RGB by removing the background of image; Finally, obtaining ideal binary image through threshold segmentation and mathematical morphology processing on image, and realizing geometric features extraction of the flame image. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|