Plant Species Identification Using Computer Vision Techniques: A Systematic Literature Review

Area of Science:

• Botany
• Computer Vision
• Biodiversity Conservation

Background:

• Traditional plant identification is complex and time-consuming for non-experts.
• Automating species identification is increasingly important for biodiversity research.
• Advancements in digital technology and image processing enable automated identification.

Purpose of the Study:

• To systematically review and compare computer vision approaches for plant species identification.
• To analyze primary studies published between 2005 and 2015.
• To provide a guide for researchers and beginners in the field.

Main Methods:

• Systematic literature review of 120 peer-reviewed studies.
• Categorization of methods by plant organ and features (shape, texture, color, margin, vein structure).
• Comparison of methods based on classification accuracy on public datasets.

Main Results:

• Analysis of various computer vision techniques applied to plant identification.
• Comparison of method performance based on accuracy metrics.
• Identification of trends and common approaches in the field.

Conclusions:

• Computer vision offers a promising avenue for automating plant species identification.
• This review provides a comprehensive overview and comparison of existing methods.
• The findings aid researchers and novices in selecting appropriate identification techniques.