Early tobacco virus disease detection method based on infrared thermal imaging technology

Abstract

The invention discloses an early tobacco virus disease detection method based on an infrared thermal imaging technology, which comprises the steps that a tobacco infrared thermal image is subjected to image intensification from image smoothing, image sharpening and histogram equalization; a new tobacco leaf thermal image area and an old tobacco leaf thermal image area are extracted by a threshold segmentation method; if a temperature of a growth environment of tobacco is higher than 16 DEG C and less than or equal to 28 DEG C, a difference value between temperature values of various pixel points in the new tobacco leaf area and an average value of the temperature values is calculated; when an absolute value of the difference value is greater than 1 DEG C, a plant is a disease plant; if not, the plant is not the disease plant; when the environment temperature is higher than 28 DEG C and less than 34 DEG C, a difference value between average temperatures of the pixels in the new tobacco leaf area and the old tobacco leaf area is calculated; when the difference value is less than 0.8 DEG C, the plant is the disease plant; and if not, the plant is not the disease plant. The method is simple in step, short in consumed time and high in accuracy, can detect a disease earlier, can be applied to long-distance real-time monitoring of a field crop, and provides timely and reliable information of a crop growth condition, a disease condition and the like.

Description

technical field [0001] The invention relates to an early detection method of tobacco virus disease, in particular to a method for early detection of tobacco virus disease by using infrared thermal imaging technology, and belongs to the field of non-destructive detection. Background technique [0002] Tobacco is an important economic crop and an important source of income for the state and local taxation, and is highly valued by relevant departments. In recent years, tobacco virus disease has occurred in various countries in the world, and outbreaks often occur, causing huge losses to tobacco production and directly dampening the enthusiasm of tobacco farmers. In my country, the occurrence area and damage degree of tobacco virus disease have increased significantly. The rate of diseased plants in the field is generally 20%-40%, and the serious one is 40%-80%. Some plots even have no production. After the tobacco is infected with the virus, the growth of the leaves is inhibit...

AI Technical Summary

Problems solved by technology

After the tobacco is infected with the virus, the growth of the leaves is inhibited, the leaves are small and deformed, and the yield reduction rate can reach 20%-80%. At the same time, the color of the diseased leaves is uneven, the taste is poor, and the quality declines. At present, there is no treatment A drug with a relatively ideal effect, an out-of-control viral disease disaster is enough to destroy a smoking area, so early diagnosis and early detection of diseased plants are particularly important

[0003] At present, immunological methods such as enzyme-linked immunosorbent reaction and rapid detection test strips are commonly used in tobacco virus detection. This method has the advantages of high sensitivity, strong specificity and easy operation, but it has its limitations and Insufficient - various tobacco viruses have different physical and chemical properties, morphological properties and antigenic properties, so antibody production, determination and preparation of enzyme markers cannot be standardized. In addition, specific and non-specific interference will also be encountered during detection

In actual planting, tobacco farmers judge the disease based on experience and symptoms of diseased leaves, that is, the artificial sensory evaluation method. This method is highly subjective and unreliable, and is often based on the obvious and typical symptoms of the disease on the plant. It is too late carry on

[0004] Existing virus disease detection technology has disadvantages of cumbersome steps, damage, and not suitable for large-scale detection in the field.