Rough set theory-based foundation cloud picture cloud classification and identification system and method

Abstract

The invention belongs to the field of image processing, and aims to accurately identify clear sky, thin cloud and thick cloud, prepare for calculating the output condition of the current photovoltaicpower generation, and predict the output of the photovoltaic power generation in a short period by combining the motion condition of the cloud. The invention discloses a rough set theory-based foundation cloud atlas cloud classification and identification system and method. The system comprises a camera, a hemispherical mirror surface with a heating device, a shading belt above the mirror surfaceand a computer; a camera above the instrument is used for vertically and downwards shooting a hemispherical mirror surface with a heating device to obtain an image presented in the sky at the moment,the shot image is automatically stored in a computer, a rough set module is set in the computer, a threshold value is solved by utilizing a rough set model, and pixels of the image are divided into different areas, thereby carrying out classification and identification on the sky mode. The system and method are mainly applied to photovoltaic power generation occasions.

Description

technical field [0001] The invention belongs to the field of image processing, in particular to a method for processing ground-based cloud images. Background technique [0002] At present, cloud recognition methods are mainly divided into two categories: traditional image processing methods and artificial intelligence methods. Among them, although the traditional methods have a high accuracy rate, their scope of application is often limited, and corresponding models need to be established according to specific problems; while some deep learning algorithms in artificial intelligence methods have a wide range of applications, but they are difficult to identify complex shapes such as cloud images. Images with many details are not very accurate, and it is difficult to meet the accuracy requirements for short-term forecasting of photovoltaic output, and a large number of tedious full-image calibrations are required. In addition, most of the existing methods only divide the cloud...

AI Technical Summary

Problems solved by technology

Among them, although the traditional methods have a high accuracy rate, their scope of application is often limited, and corresponding models need to be established according to specific problems; while some deep learning algorithms in artificial intelligence methods have a wide range of applications, but they are difficult to identify complex shapes such as cloud images. Images with many details are not very accurate, and it is difficult to meet the accuracy requirements for short-term forecasting of photovoltaic output, and a large number of tedious full-image calibrations are required

In addition, most of the existing methods only divide the cloud image into two modes: cloud and non-cloud, which indeed meet the needs of many other problems, but cannot meet the accuracy requirements of photovoltaic forecasting.

Images