Guard net detection method based on Fourier spectrum under polar coordinates

Abstract

The invention discloses a guard net detection method based on the Fourier spectrum under polar coordinates. The detection method comprises the following steps that S1) a guard-net training sample library is established; S2) sliding window is carried out on images in a major sample image library, and Fourier transform is carried out on each image after sliding window; S3) polar coordinate transformation is carried out on the images after Fourier transform, and peak values are used as texture features of the guard net, S4) the extracted texture features are transmitted into a classifier to obtain an intelligent classifier based on the guard-net training sample library; and S5) guard net detection after a training phase is completed. According to the method, guard net detection is used as basis for determining illegal construction, a computer is combined with a camera to automatically monitor state-owned land and automatically identify guard nets to determine whether illegal construction occurs, the instantaneity and timeliness are high, discovery of illegal construction is not lagged, monitoring is stable and uninterrupted, consumed energy and material are low, and time and labor are saved.

Description

technical field [0001] The invention relates to the technical field of image processing and image recognition, in particular to a protective net detection method based on Fourier spectrum under polar coordinates for land supervision and management. Background technique [0002] As we all know, the focus of the land supervision and management department is to crack down on illegal occupation of state-owned land and effectively ensure the safety of state-owned reserve land. In the prior art, the monitoring methods are manual inspections periodically or irregularly to prevent private units or individuals from illegally occupying state-owned land and building illegally, which will consume huge manpower, financial resources and material resources At the same time, for state-owned land in remote mountainous areas and other places, due to the inability to achieve real-time supervision and monitoring, it also makes it difficult to discover private units or individuals illegally occu...

AI Technical Summary

Problems solved by technology

In the prior art, the monitoring method is to conduct periodic or irregular inspections manually to prevent illegal occupation and illegal construction of state-owned land by private units or individuals, which will consume huge manpower, financial and material resources At the same time, for state-owned land in remote mountainous areas and other places, real-time supervision and monitoring cannot be achieved, which also makes it difficult to discover private units or individuals illegally occupying state-owned land and illegal construction on state-owned land, and the existing situation also exists The narrow monitoring area and other issues make the monitoring weak, and the state-owned land resources cannot be managed efficiently and in real time

Images