Citrus Leaf Disease Detection using Joint Scale Local Binary Pattern

  • Madiha Ashfaq
  • S. M. Adnan Shah
  • Wakeel Ahmad
  • Shahbano
  • Muhammad Ilyas


Plant diseases are known from early times of cultivation and is also considered as one of the major factors affecting crops growth and quantity. Such losses from plant diseases can have a substantial economic impact, causing a reduction in income of farmers and higher prices for consumers. Today’s agriculture demands regular use of high-tech technologies such as robots, moisture sensors, aerial surveillance and images processing. These technological shifts allow farmer to be more profitable and more environmentally friendly. This paper investigates and present Joint Scale Local Binary Pattern (JS-LBP) algorithm for detection of plant leaf disease based on image segmentation and texture analysis. The proposed method consists of two main phases, identification of disease lesion spot on citrus leaves and its classification. Leave images are pre-processed using top-hat filtering and two median filters, later these images are fed into the JS-LBP descriptor for feature extraction. Experiment shows that proposed algorithm attains 98.6% accuracy rate. Confusion matrix and ROC curve for different citrus diseases are also presented.

How to Cite
Madiha Ashfaq, S. M. Adnan Shah, Wakeel Ahmad, Shahbano, & Muhammad Ilyas. (2020). Citrus Leaf Disease Detection using Joint Scale Local Binary Pattern. University of Sindh Journal of Information and Communication Technology , 4(4), 199 - 206. Retrieved from

Most read articles by the same author(s)

Obs.: This plugin requires at least one statistics/report plugin to be enabled. If your statistics plugins provide more than one metric then please also select a main metric on the admin's site settings page and/or on the journal manager's settings pages.