Transformer nameplate information acquisition method and intelligent acquisition system

Abstract

The invention discloses a transformer nameplate information acquisition method. The method comprises the following steps: acquiring a transformer nameplate image by using camera equipment; segmentingthe transformer nameplate image on a transformer image acquisition site by adopting a lightweight neural network image recognition program to segment small character pictures in the transformer nameplate image; using a PCAnet network computer program to recognize the small character pictures in the step 2, and converting the small character pictures into characters; and registering the transformernameplate information obtained in the step 3. The invention also discloses two transformer nameplate information intelligent acquisition systems. According to the invention, after on-site image acquisition of the transformer nameplate, nameplate information identification is divided into two steps, so that transmitted image data is greatly reduced, character identification is more efficient, a background character identification result is returned to a transformer image acquisition site, the character identification result is convenient to check, and the acquired transformer nameplate information is ensured to be accurate and errorless.

Description

technical field [0001] The invention belongs to the field of transformer nameplate identification, and in particular relates to an intelligent collection system for distribution transformer nameplate information. Background technique [0002] The information collection of distribution transformers is the primary task to realize the informatization of power grid. Accurate and convenient acquisition of transformer nameplate information is required. The transformer is usually 3 meters above the ground, and it is difficult to see the information on the nameplate of the transformer with the naked eye. Due to the viewing angle and light, it is difficult to obtain more accurate information on the nameplate even with a SLR camera and other tools. [0003] In terms of nameplate recognition, the traditional OpenCV-based nameplate positioning effect is remarkable, but subsequent character template matching and recognition are required, and the operation is cumbersome. Now the neural ...

AI Technical Summary

Problems solved by technology

The transformer is usually 3 meters above the ground, and it is difficult to see the nameplate information of the transformer with the naked eye. Due to the observation angle and light, it is difficult to obtain more accurate nameplate information even with tools such as SLR cameras.

[0003] In terms of nameplate recognition, the traditional OpenCV-based nameplate positioning effect is remarkable, but subsequent character template matching and recognition are required, and the operation is cumbersome.

Now the neural network has a remarkable effect in the field of image character recognition. By adding a trained CNN network model to the device software, the high accuracy of nameplate information recognition can be achieved. However, training a better CNN network model takes a lot of time and samples, and The structure of Chinese characters is complex, the sample library is huge, the calculation scale will also increase, and the time complexity of the algorithm will increase significantly, which is not suitable for running on mobile portable devices

[0004] Therefore, research on an information collection tool and an efficient recognition algorithm solves the problem that the transformer nameplate information is physically difficult to collect and the structural information of the transformer nameplate is difficult to obtain

Images