← Back to Publications List

Potato Disease Classification Using Densent121

Students & Supervisors

Student Authors
Saptanil Ghose
Bachelor of Science in Computer Science & Engineering, FST
Supervisors
Firoz Ahmed
Professor, Faculty, FST

Abstract

Image processing-based approaches for plant disease recognition and classification represent an important area of current study. Such applications enable real-time detection of diseases in plants. Plants remain vulnerable to fungal, bacterial, and viral infections. Most existing studies consider only a limited range of potato diseases and are performed under controlled conditions. However, robust real-time systems capable of addressing diverse environments, unseen diseases, and variations in image quality remain lacking. This study considered three potato conditions: Early Blight, Late Blight, and Healthy Potato Plant Leaves. Color, shape, and texture features extracted from both healthy and diseased potato plant images were applied for classification. The feature extraction process followed the segmentation step, and features obtained from segmented images were then used as inputs to the classification model. Finally, three types of classifications were employed to recognize disease presence. Using these three potato leaf categories resulted in an overall classification accuracy of 98%. Previous research has also attempted to develop more generalized frameworks for potato disease detection

Keywords

Potato Dieses

Publication Details

  • Type of Publication:
  • Conference Name: 3rd International Conference on Big Data, IoT and Machine Learning (BIM 2025)
  • Date of Conference: 25/09/2025 - 25/09/2025
  • Venue: Dhaka International University (DIU)
  • Organizer: Department of CSE and EEE, DIU