Texture entropy value-based fan blade identification method and system

Abstract

The invention discloses a texture entropy value-based fan blade identification method and system, and belongs to the technical field of machine vision. Firstly, initial image information of fan blades is obtained, the texture entropy value of each blade is calculated through an image processing technology, the distribution condition of the texture entropy values in the blades is recorded, and a change threshold value of the texture entropy values in unit time is estimated according to meteorological information of a to-be-recognized image in a period of time near a collection moment; and comparing and identifying the texture entropy value of the to-be-identified target in the newly collected image and the texture entropy value of the corresponding region in the initial image by superimposing the change threshold value. According to the method, the fan blade can be automatically identified, so that faults and defects can be quickly positioned later, and the quick development of later maintenance work is ensured; meanwhile, automatic blade identification can be carried out under the condition that the fan does not need to stop, the stop inspection time of the wind turbine generator is shortened, the automatic inspection efficiency is improved, the power generation time of the wind turbine generator is shortened, the power generation efficiency of the wind turbine generator is improved, and good application prospects are achieved.

Description

technical field [0001] The invention belongs to the technical field of machine vision, and in particular relates to a method and system for identifying a fan blade based on a texture entropy value. Background technique [0002] In the daily operation and maintenance process of wind turbines (referred to as fans), due to the scattered distribution of wind turbines, long distances between each other, and wide distribution area, manual inspection is difficult, inefficient, and has certain risks. For long-point, multi-scope and large-scale inspection scenarios, the labor cost is also high. Therefore, in order to maintain the efficiency and speed of the wind turbine inspection process, it is necessary to use automated inspection equipment, including but not limited to drone cameras. The condition of the fan unit, especially the blades, is obtained, and the automatic image processing method is used to detect whether there are faults such as cracks in the blades. [0003] The use ...

AI Technical Summary

Problems solved by technology

[0002] During the daily operation and maintenance of wind turbines (referred to as wind turbines), due to the scattered distribution of wind turbines, the distance between them is long, and the distribution area is wide, manual inspection is difficult, inefficient, and dangerous. For long-term, multi-scale and large-scale inspection scenarios, the cost of using manual methods is also high. Therefore, in order to maintain the efficient and fast inspection process of wind turbines, it is necessary to use automated inspection equipment, including but not limited to drone cameras. Obtain the conditions of the fan unit, especially the blades, and detect whether there are cracks and other faults in the blades through automated image processing methods

[0003] Although the automatic image processing method for fault detection of fan blades has the advantages of high efficiency and safety, however, during the fault detection process, fan blades need to be photographed from different angles, and it is difficult to automatically identify each blade during the shooting process, which may cause In addition, when the fan is inspected without stopping, the positions of the blades in the images taken at different times may change. At this time, even if the problem on the blade is found, it cannot Determine which blade has the problem, which brings inconvenience to the defect analysis and discovery of fan blades

Images