Abstract
Nanomaterials are used in almost every engineering field. The properties of synthesized nanomaterials are greatly influenced by synthesis techniques and conditions. Digital image processing techniques are critical in identifying the size and structure of images, and precisely classifying them allows scientists and researchers to use them in a variety of applications. The benefits of digital image processing techniques improve object recognition accuracy in computer vision and pattern recognition. The primary focus of any image processing research is the image that needs to be cleaned up to examine the expressions or features in it. The majority of images have noise in them, either from natural occurrences or from the data collection process. Pre-processing the images enhances the image's quality, which facilitates the subsequent image identification procedure. The present study primarily focuses on the extraction of various geometrical features from copper oxide(CuO) FESEM nanoparticle images, which uses multi-level thresholding by Otsu’s method for segmentation. Further, for the extraction of pores, the watershed transform technique is applied and geometrical features such as; pore size(area), pore radius, porosity, average porosity, and overall average pore radius, and average pore size(Area) are computed. The CuO FESEM nanoparticle images are captured at varying magnifications namely; image A at 2 KX, image B at 5 KX, image C at 10 KX, image D at 20 KX, and image E at 100 KX of magnification have been used. The algorithms show that; as the magnification of images increases the porosity decreases. The standard porosity of CuO FESEM nanoparticles ranges between 0.12% to 0.15%. The proposed results show a porosity of 0.13%. The toxic effect of CuO nanoparticles is size dependent. The smaller the CuO nanoparticle size, the more toxic. As the size of nanoparticles increases the toxicity is less. CuO nanoparticles above 80 nm size show less toxicity than those below 80 nm. The proposed study computes the size of CuoO nanoparticles that range between 90 nm to 400 nm, and hence are comparatively nontoxic in nature. Furthermore, the results are manually verified with chemical experts, demonstrating the exhaustiveness of the proposed method
Keyword
Nanomaterials, Nanoparticle, Copper Oxide(CuO), FESEM, Magnification, Image segmentation, Porosity
PDF Download (click here)
|