Image Fusion Technology Application in Water Quality Monitoring Based on Digital Microscopic

Abstract:

In the process of water quality monitoring based on microscopic examination of the activated sludge, low-contrast images can be captured sometimes when manual observation is being replaced by automated observation based on machine vision. It is difficult to accurately detect the target complete edge in a low-contrast image with a single image-segmentation method. The idea provided is in the paper that the low-contrast image is firstly segmented by Canny operator and mathematical morphology respectively, and then the two segmented images are fused based on wavelet transform. In order to verify the idea, the tail spines insect image shot by digital microscope is processed according to the idea in the paper. The result shows that this idea can achieve better image edge detection effect and it will provide good technical basis for the next protozoan and metazoan identification and statistics.

• Periodical: Applied Mechanics and Materials (Volumes 295-298)
• Pages: 909-914
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.295-298.909
• Online since: February 2013
• Authors: Yi Hong Wang, Shi Feng Tao, Jun Chen
• Keywords: Low-Contrast Image, Microscopic Examination, Targets Extraction, Wavelet Transform (WT)

References

[1] Raghav, A., Hasan, R., Mahmood, S.: History of Non Artificial Intelligence Based Biological Monitoring of River Water Quality, ICIII (2011), Vol. 3, pp.248-253.

DOI: 10.1109/iciii.2011.342

[2] Hasan, R., Mahmood, S., Raghav, A., Hasan, M.A.: Artificial Intelligence in River Quality Assessment, ICIII (2011), Vol. 1, pp.491-495.

[3] Ashish Tripathi, Jabbour, Rabih E., etc.: Detection and identification of a water mixture of E. coli cells and B. subtilis spores with raman chemical imaging microscopy, QELS (2007), pp.1-2.

DOI: 10.1109/qels.2007.4431543

[4] Xinyuan Liu, Can Wang, etc.: Isolation and Characteristics of Degradation Strains of Oil Field Wastewater, RSETE (2012), pp.1-4.

[5] CesarA. B: Biologieal shaPe characterization for automatic image reeognition and diagnosis of Protozoan Parasites of genus Eimerial [J], Pattern Recognition, (2007), (40), p.1900-(1910).

DOI: 10.1016/j.patcog.2006.12.006

[6] Y.P. Ginoris: Recognition of Protozoa and metazoan using image analysis tools, diseriminant analysis, neural networks and decision trees[J], Analytica Chimica Acta, (2007), pp.160-169.

DOI: 10.1016/j.aca.2006.12.055

[7] Hasikin, K., Isa, N.A.M.: Enhancement of the Low Contrast Image Using Fuzzy Set Theory, UKSim 14th International Conference on (2012), pp.371-376.

DOI: 10.1109/uksim.2012.60

[8] Hsiao-Chiang Chuang, Comer, M.L.: A new method for segmentation of noisy, low-contrast image sequences, ISCAS ( 2010), pp.2868-2871.

[9] Zhang Yingjie, Ge Liling: New Approach to Low Contrast Image Segmentation, ICBBE(2008), pp.2369-2372.