Substation instrumentation target detection based on multi‐scale feature fusion

Abstract

SUMMARY With the promotion of smart grid construction work, the use of high‐precision and high‐efficiency substation inspection robot has become the development trend of substation inspection. A multi‐scale feature fusion meter target detection algorithm is proposed to address the problems of low efficiency and susceptibility to surrounding environmental factors by the traditional manual meter reading method. Kinecct is used to acquire color images of substation meters with different backgrounds, light intensities, and angles to build a substation meter dataset. Based on the complementarity and correlation of multi‐scale features, an SSD target detection model with multi‐scale feature fusion is established, and the performance of the algorithm is tested on the constructed dataset, and comparative experiments are conducted to verify the effectiveness of the algorithm for target detection accuracy improvement.

Improved single shot detection using DenseNet for tiny target detection

“…Currently, the common improvement methods are to improve on the mainstream models and design them specifically for the feature characteristics of small targets. Based on the difference of ideas, the approaches can be broadly classified into multi-scale feature prediction, [39][40][41][42][43] improving feature resolution, 44,45 extracting contextual information, 46,47 designing backbone networks and training strategies. [48][49][50][51] Deep learning has been shown to have excellent performance, [52][53][54][55][56][57][58] and an increasing number of researchers have launched studies on target detection based on this.…”

“…Currently, the common improvement methods are to improve on the mainstream models and design them specifically for the feature characteristics of small targets. Based on the difference of ideas, the approaches can be broadly classified into multi‐scale feature prediction, 39‐43 improving feature resolution, 44,45 extracting contextual information, 46,47 designing backbone networks and training strategies 48‐51 …”

Multi-Objective Location and Mapping Based on Deep Learning and Visual Slam

“…On the one hand, the mapping and positioning of the classical SLAM system are mostly based on the geometric matching at the pixel level, lacking a priori information to help the sensor locate the objects [ 17 , 18 ]. Data association can be improved from the traditional pixel level to the object level with the help of semantic information, which can help improve the positioning accuracy in complex scenes [ 19 , 20 , 21 ]. Compared with the traditional physical geometric information, semantic information has consistency and stability and is less affected by environmental transformations.…”

Large scale instance segmentation of outdoor environment based on improved YOLACT

“…[1][2][3][4] The image often contains a lot of valuable information, and the purpose of segmentation is to simplify or change the representation of the image, which makes the image easier to understand and analyze. [5][6][7][8][9] Because of developments in information and communication technology, the instance segmentation assigns a category or other information to each pixel in an image, and pixels with the same information mean that such pixels have similar or identical visual characteristics. [10][11][12] With the gradual popularization of hardware devices for image acquisition and the development of the Internet, the process of human cognition of images is transformed into the process of computer cognition of images, and there are more and more applications that use computers to cognize and process images.…”

“…As the main carrier of visual information, images play an important role in perception such as target detection and target recognition, which is currently an important branch within the hottest AI areas 1–4 . The image often contains a lot of valuable information, and the purpose of segmentation is to simplify or change the representation of the image, which makes the image easier to understand and analyze 5–9 . Because of developments in information and communication technology, the instance segmentation assigns a category or other information to each pixel in an image, and pixels with the same information mean that such pixels have similar or identical visual characteristics 10–12 .…”