Transmission line bird's nest identification method based on hog features and machine learning

Abstract

The invention belongs to the field of electric power technology and computer vision. In order to separate normal pictures from pictures with bird's nests, locate and find problems faster, and meet the needs of smart grid construction, the invention provides a transmission line bird's nest based on HOG features and machine learning. Recognition method, steps are as follows: one, direction gradient histogram HOG feature extraction; Two, principal component analysis three, training support vector machine SVM classifier 1) normalization; 2) extract the feature vector composition of the training set gained in the previous step The training set of the classifier, and make a label file conforming to the SVM format; 3) find the best parameters through testing; 4, input the test set image, use the trained classifier to classify, and output the final classification result. The invention is mainly applied to the occasion of automatically identifying the bird's nest fault of the power equipment by using the image.

Description

technical field [0001] The invention belongs to the fields of electric power technology and computer vision, and relates to a method for identifying bird's nests of power transmission lines based on histogram of orientation gradient (HOG) features and machine learning. Specifically, it involves a bird's nest identification method for transmission lines based on HOG features and machine learning. Background technique [0002] Transmission lines play a very important role in the power system and are directly related to the electricity consumption of all walks of life in society. Large-scale power outages will bring immeasurable losses to the country's economic development. Therefore, the safe operation of transmission lines is one of the issues of great concern to the power sector. [0003] Bird activity often interferes with the normal operation of transmission lines. Its impact on transmission lines is mainly reflected in the following aspects: birds often build their nest...

AI Technical Summary

Problems solved by technology

[0005] However, this system requires long-term monitoring by staff, and is greatly affected by human factors. Long-term work will inevitably make it difficult for staff to concentrate for a long time, and work efficiency will decrease; in addition, if the subjective judgment of staff is used for these data , it is very easy to misjudge or miss the judgment, it is difficult to accurately find the potential safety hazards of power transmission equipment, and it greatly increases the maintenance cost, which cannot meet the needs of smart grid construction

Images