Rosette plant segmentation with leaf count using orthogonal transform and deep convolutional neural network
- PDF / 1,859,349 Bytes
- 14 Pages / 595.276 x 790.866 pts Page_size
- 79 Downloads / 174 Views
ORIGINAL PAPER
Rosette plant segmentation with leaf count using orthogonal transform and deep convolutional neural network J. Praveen Kumar1
· S. Dominic2
Received: 30 May 2019 / Revised: 2 October 2019 / Accepted: 16 December 2019 © Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Plant image analysis plays an important role in agriculture. It is used to record the morphological plant traits regularly and accurately. The plant growth is one of the key traits to be analyzed, which relies on leaf area (i.e., leaf region or plant region) and leaf count. One of the ways to find the leaf count is counting the leaves using segmented plant region. In this paper, a new plant region segmentation scheme is proposed in the orthogonal transform domain based on orthogonal transform coefficients. Initially, an analysis of orthogonal transform coefficients is carried out in terms of the response of orthogonal basis vectors to extract the plant region. After extracting the plant region, the L*a*b and CMYK color spaces are used for noise removal in the segmentation scheme. Finally, the leaves are counted using fine-tuned deep convolutional neural network models. The proposed scheme is experimented on CVPPP benchmark datasets and also tested with the images taken from mobile phone to ensure its reliability and cross-platform applicability. The experiment results on CVPPP benchmark datasets are promising. Keywords Deep convolutional neural network · Leaf count · Orthogonal transform coefficients · Plant segmentation
1 Introduction Plant phenotyping is vital to sustain with the increasing global demand for food. It is an important sector to enhance plant resistance and productivity. The traditional plant phenotyping involves in measuring the key traits manually, which becomes a tedious, slow and expensive task. In many cases, traditional plant phenotyping techniques use random measurement which happens to be a measurement bias. In recent days, plant phenotyping becomes a bottleneck in the field of modern research programs and plant breeding [11]. Computer vision is one of the important tools to free the bottleneck in the field of plant phenotyping [25]. Therefore, many researchers have started their research recently in image-based plant phenotyping techniques which include the research in identifying the morphological plant traits and observing the development and growth of plant by analyzing the digital plant images.
B
J. Praveen Kumar [email protected]
1
Vellore Institute of Technology (VIT-AP), Amaravathi, Andhra Pradesh, India
2
National Institute of Technology, Tiruchirappalli, India
By automating the process of estimating these visual traits to a satisfactory level, the costs can be reduced and the production speed can be improved. Leaf count and leaf area (i.e., plant region) are the key plant phenotyping traits used to analyze the plant growth and development [27,32,36], flowering time [18] and yield potential. The leaf count can be addressed in various ways [3] from the machine learn
Data Loading...
