A Novel Enhancement Algorithm for Non-Uniform Illumination Particle Image
In order to improve the influence of non-uniform illumination and the measurement precision, a novel enhancement algorithm was proposed. Although bad illumination condition infection could be removed by MSR, when used in particle images, contrast enhancem
- PDF / 463,706 Bytes
- 8 Pages / 439.37 x 666.142 pts Page_size
- 80 Downloads / 192 Views
A Novel Enhancement Algorithm for Non-Uniform Illumination Particle Image Liu Weihua
Abstract In order to improve the influence of non-uniform illumination and the measurement precision, a novel enhancement algorithm was proposed. Although bad illumination condition infection could be removed by MSR, when used in particle images, contrast enhancement effect was not been satisfactory. Then a non-linear gray transformation was introduced in image contrast extending. The experiments proved that enhanced images processed by the novel algorithm had more uniform background and higher contrast than former enhancement algorithmic. Over enhancement was avoided and it also could improve segmented efficient and ensure accurate segmented results. Keywords Non-uniform illumination Gray non-linear transformation
Particle image Multi-scale Retienx
26.1 Introduction With the development of machine vision, particle detection methods based on machine vision also have very great improvement. Then morphological parameters, which include particulate matter size, fractal dimension, shape factors and so on, can be got using image analysis. Machine vision is approach to measure micrometre particle or nanometer particle structure information, quantitative analysis of powder nature and other useful particle information, the most important is it can real-time monitor its industrial process. Make use of advantages of machine vision technique is promptly and accurately, analysis particle size
L. Weihua (&) School of Management Science and Engineering, Shandong University of Finance and Economics, 250014 Jinan, Shandong, China e-mail: [email protected]
S. Li et al. (eds.), Frontier and Future Development of Information Technology in Medicine and Education, Lecture Notes in Electrical Engineering 269, DOI: 10.1007/978-94-007-7618-0_26, Springer Science+Business Media Dordrecht 2014
265
266
L. Weihua
distribution, number concentration, and weight concentration which has important role in guiding and monitoring industrial process. Once the images have been obtained, digital image processing techniques can be applied to extract information such as particle size distribution and absolute solids concentration and so on. The main processing challenge lies in separating the particles from the background in a consistent and repeatable manner— inconsistent image segmentation will lead to variances in measured particle size and, since other measurements such as concentration are simply arrived at through statistical analysis of particle size data, the variances will lead to cascading measurement errors [1].The simplest form of image segmentation is thresholding. This method is fast and easy to implement but, in order to provide good result, a very high contrast between particles and background must be present in the images. The quality of particle images is poor because of non-uniform illumination, and the segmentation result is also unsatisfactory when using classical greyimage enhancement algorithm such as histogram equalization. So a ne
Data Loading...
