Flotation Bubble Delineation Based on Harris Corner Detection and Local Gray Value Minima

Abstract

Abstract:Froth image segmentation is an important and basic part in an online froth monitoring system in mineral processing. The fast and accurate bubble delineation in a froth image is significant for the subsequent froth surface characterization. This paper proposes a froth image segmentation method combining image classification and image segmentation. In the method, an improved Harris corner detection algorithm is applied to classify froth images first. Then, for each class, the images are segmented by automatically choosing the corresponding parameters for identifying bubble edge points through extracting the local gray value minima. Finally, on the basis of the edge points, the bubbles are delineated by using a number of post-processing functions. Compared with the widely used Watershed algorithm and others for a number of lead zinc froth images in a flotation plant, the new method (algorithm) can alleviate the over-segmentation problem effectively. The experimental results show that the new method can produce good bubble delineation results automatically. In addition, its processing speed can also meet the online measurement requirements.

Building YoloV4 models for identification of rock minerals in thin section

“…Image recognition has been widely developed in the last decade using deep learning methods [5]. Studies have also been carried out to introduce minerals in the form of images [6]. Image development using deep learning methods will be influenced by the number of mineral datasets [7].…”

Rock Particle Image Segmentation Based on Improved Normalized Cut

“…They are: Crashed particles (aggregates) from a falling stream; Blasted rock particles (fragments) from muckpiles; and natural rock particles (aggregates). The algorithm proposed in the paper is used in these particle images, and it is compared with some widely used methods, such as Auto-threshold segmentation [3,[6][7], Contour extraction based on priori knowledge [26][27][28][29][30], Minimum spanning tree segmentation [21], FCM segmentation [11][12], Clustering segmentation [10] and Watershed segmentation [8][9]. The testing results show that the improved Normalized Cut is suitable both for densely packed rock particle images and sparsely distributed particle images, and the segmentation results are satisfactory.…”

“…In recent years, with the development of computer science, the image processing technology is gradually applied into the different research fields of the rock particle image processing [2][3][4][5]. Currently, Threshold method [6][7], Watershed [8][9], Edge detection [7], Clustering analysis [10], Fuzzy C-Means [11], Fuzzy clustering analysis methods [10][11] and Graph based algorithms such as Minimum spanning tree [12][13][14][15][16][17][18][19][20][21] are often studied and used for rock particle image segmentation, but these methods are mainly based on gray level information, not on considering the combination of other information.…”

Froth Image Acquisition and Enhancement on Optical Correction and Retinex Compensation

“…Hence, many scholars have focused on the froth image acquisition system in recent years. In 1997-2000 [2,3], the European Union invested a large amount of money launched the EU project "bubbles results based on machine vision and color characterization", and one of the main research goals was to setup an image acquisition system, where the researchers contacted different industrial companies to get a high-quality and low-cost froth image acquisition system, and the froth is mainly for copper minerals. The scholars in Chile and in South Africa also applied machine vision systems to monitor platinum, graphite, and copper flotation [2][3][4][5], and they have the same problems as the above system for acquiring high quality images, because the flotation materials are different despite the image acquisitions being different.…”