Mineral flotation froth image texture analysis and working condition identification method based on Minkowski distance

Abstract

The invention discloses a mineral flotation froth image texture analysis and working condition identification method based on the Minkowski distance. The method includes the steps that firstly, froth image samples of mineral froth flotation under different working conditions are acquired and preprocessed; then, the preprocessed froth image samples are segmented through a watershed segmentation algorithm, each froth size is counted and acquired so that parameters in a Minkowski distance formula can be determined, a complex network model of froth images is established, and the energy and the entropy of the model are calculated and serve as froth image texture description indexes; finally, the acquired froth image texture description indexes serve as feature vectors of the image samples and a linear discriminant classifier is trained, so that the tested image samples are classified, and the working condition in the rear-time flotation process is identified. According to the method, calculation is simple, classification is high in accuracy, and the method can be used for real-time monitoring of the texture feature extraction, the classification identification and the flotation process working condition of the mineral flotation froth images.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to a mineral flotation foam image texture analysis and working condition recognition method based on Minkowski distance. Background technique [0002] With the rapid development of computer technology, computer vision has been widely used in process monitoring and optimal control of mineral froth flotation. As the key vision of the flotation foam surface, the texture can not only directly reflect the working conditions of the flotation production process, but also serve as a direct indicator of the production process indicators, which is an important basis for predictive control of the flotation working conditions based on computer vision. Therefore, accurately characterizing the texture characteristics of the foam image and guiding the optimal control of the flotation process play a very important role in reducing the production cost of enterprises and improving economic benefits. ...

AI Technical Summary

Problems solved by technology

Due to the microscopic heterogeneity, complexity, and ambiguity of the foam texture, although researchers have proposed many texture feature extraction methods in recent years, such as gray-level co-occurrence matrix method, wavelet transform method, fuzzy texture spe

Images