.png){kind=icon}
.png){kind=icon}
Return
On solving leaf classification using linear regression
.png){kind=icon}
PRE
AI
DOI：10.1007/s11042-020-09899-y.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Plant's conservation is getting close attention nowadays. It requires awareness about ecology among masses. Plant species identification has been proved as a primary step in literature for biodiversity conservation. It is a sequential process from leaf images as input followed by image enhancement algorithms, and feature extraction phase to classification. The complete process of identifying a leaf image requires substantial time. The article focuses on introducing a simpler and computationally inexpensive framework with a performance at par or better as compared to the existing framework. The article covers several findings and results while transforming the proposed framework for plant identification to a parameter specific optimized framework. The findings include optimizing the leaf image dimension, the impact of RGB to grayscale conversion method, and comparative analysis of the proposed framework for classification from images with other frameworks that first extract specific features and then classify. It also represents the whole framework as a regression problem. Further, improvement is incorporated by integrating the benefits of kernel trick in linear regression. Our finding confirms that the framework not only recognizing the leaf images with comparable accuracy but also reduces the computational time significantly to identify leaf images as compared to other frameworks.
Keywords:
Linear regression
Kernel function
Color to gray-scale conversion
Image down-sampling
Image projection
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
3
Papers:
1.9W
Citations:
3.2W