Identification and localisation of multiple weeds in grassland for removal operation

Wang, Jinjin, Yaob, Xiaopeng and Nguyen, Bao Kha (2022) Identification and localisation of multiple weeds in grassland for removal operation. The 14th International Conference on Digital Image Processing (ICDIP 2022), Wuhan, China, 20th - 23rd May 2022. Published in: SPIE Conference Proceedings. SPIE Digital Library ISBN 9781510646001 (Accepted)

[img] PDF - Accepted Version
Download (1MB)

Abstract

Weeds are a common issue in agriculture. Image-based weed identi�cation has regained popularity in recent
years as computing power increases. Researchers have successfully applied weed detection in the crop �feld and have combined the sensor (e.g.camera) and mechanical such as robotic weeders to get the location of the weeds. Meanwhile, many studies also have been conducted on the two classi�cations between grass and weed. However, there is no excellent and comprehensive weed dataset in reality because weeds are always similar and diffcult to obtain by non-specialists. Moreover, it is challenging to identify weeds from grasslands for their similar colors, sizes, and shapes. In this paper, our goal is to build a natural and effective dataset to train the classi�fier by extracting features of weeds so that weeds can be accurately identi�fied and quickly located. We investigate three weeds (Bitter Gentian, Hawk's Beard, Pedunculate) relatively common in grasslands. Then, we select the typical grassland dominated by the above weeds for data collection. A natural and effective dataset is built and has generality in the scene of actual grassland. Secondly, we extract image features, including Color, Histogram, and orientation gradient histogram(HOG), and make various combinations to accurately and comprehensively reflect the actual characteristics of weeds. Thirdly, we propose a "core zone" algorithm to locate the weeds. The algorithm mainly adopts technology in image processing, such as threshold segmentation and morphological transformations. Experiments show that our binary classifi�er is more accurate than the comparison method, and the accuracy of the multi-classifi�er is also high. In addition, the algorithm for weeds location is more efficient than the comparative method.

Item Type: Conference Proceedings
Keywords: Weed, Grassland, SVM, HOG, Classi cation, Object detection
Schools and Departments: School of Engineering and Informatics > Engineering and Design
SWORD Depositor: Mx Elements Account
Depositing User: Mx Elements Account
Date Deposited: 06 Jun 2022 13:14
Last Modified: 21 Jul 2022 14:48
URI: http://sro.sussex.ac.uk/id/eprint/106236

View download statistics for this item

📧 Request an update