Affiliation:
1. School of Information & Computer Science, Anhui Agricultural University, Hefei 230036, China
2. Anhui Provincial Key Laboratory of Smart Agricultural Technology and Equipment, Hefei 230036, China
Abstract
This paper discusses a semantic segmentation framework and shows its application in agricultural intelligence, such as providing environmental awareness for agricultural robots to work autonomously and efficiently. We propose an ensemble framework based on the bagging strategy and the UNet network, using RGB and HSV color spaces. We evaluated the framework on our self-built dataset (Maize) and a public dataset (Sugar Beets). Then, we compared it with UNet-based methods (single RGB and single HSV), DeepLab V3+, and SegNet. Experimental results show that our ensemble framework can synthesize the advantages of each color space and obtain the best IoUs (0.8276 and 0.6972) on the datasets (Maize and Sugar Beets), respectively. In addition, including our framework, the UNet-based methods have faster speed and a smaller parameter space than DeepLab V3+ and SegNet, which are more suitable for deployment in resource-constrained environments such as mobile robots.
Funder
Ministry of Agriculture, China
Anhui Provincial Key Laboratory of Smart Agricultural Technology and Equipment
Subject
Electrical and Electronic Engineering,Biochemistry,Instrumentation,Atomic and Molecular Physics, and Optics,Analytical Chemistry
Reference29 articles.
1. Image segmentation via adaptive K-mean clustering and knowledge-based morphological operations with biomedical applications;Chen;IEEE Trans. Image Process.,1998
2. Fully automatic segmentation of the brain in MRI;Atkins;IEEE Trans. Med. Imaging,1998
3. The Watershed Transform: Definitions, Algorithms and Parallelization Strategies;Roerdink;Fundam. Inf.,2000
4. Twaakyondo, H.M., and Okamoto, M. (1995, January 14–16). Structure analysis and recognition of mathematical expressions. Proceedings of the Third International Conference on Document Analysis and Recognition—Volume 1: IEEE Computer Society, Montreal, QC, Canada.
5. Layered Object Models for Image Segmentation;Yang;IEEE Trans. Pattern Anal. Mach. Intell.,2012