Plant biotechnology is an important field due to food and nutrition requirements for humans and animals. However, a significant quantity of crop loss occurs every year due to stresses. To manage this, there is different research going on for plants such as stress analysis, phenotyping, genetic improvement, nutrient & mineral deficiency and toxicity analysis, weeds protection, insect control analysis, etc. For stress analysis, different types of sensors are used for plant data collection such as temperature, humidity, camera, spectral camera, etc. The visible range is widely utilized due to camera sensor availability, but it has limitations that images can be collected for analysis only when symptoms are visible. However, another non-visible spectrum (Thermal Infrared) is becoming more useful, as it can provide plant insight information before symptoms are visible. Infrared spectrum is used for disease analysis and is able to predict disease in the presymptomatic stage before the appearance of disease symptoms in the visible spectrum. Thermal infrared imaging is utilized for machine and deep learning models for disease identification. Further, combining thermal infrared imaging with RGB imaging provides more information for disease analysis in the real field. This paper summarized disease analysis methods based on infrared imaging using image processing and deep learning.