The excellent functions of thermal imaging cameras open the door to various applications. Each research field has specific requirements such as resolution, field of view, thermal sensitivity, price or camera size. Therefore, it is foreseeable that in the next few years, the diversity of available cameras will further expand, not limited to high-end cameras.
Infrared thermal imaging has been found to be useful for two different types of problems: analysis of known objects and detection of unknown objects. In the first question, the object and its position in the image are known, and the properties of the object can be analyzed. The result may be the type, condition, or health of the substance. The methods used here usually simply record the temperature or even manually check the image. If computer vision methods are used, they are usually just simple algorithms such as threshold and blob detection. For the second question, the type of object or its position in the image is unknown. The most important step in this type of problem is usually the detection and classification of objects. The goal here is usually to design an automated system, for example to detect or track specific objects. More advanced computer vision algorithms can be applied here to design powerful and automated systems. In applications where the temperature of the subject is different from the surrounding environment, compared with visual cameras, infrared thermal imaging cameras can greatly simplify the detection steps.
Due to the lower price and higher availability of thermal imaging cameras, methods for analyzing known objects and detecting unknown objects are rapidly expanding. In the case of known objects, thermal imaging cameras can be regarded as alternatives to non-contact thermometers. In the last case, thermal imaging cameras are more regarded as alternatives to vision cameras. Therefore, from the perspective of computer vision, thermal imaging cameras are currently receiving more and more attention. However, the general trend in modern society is to implement automation. With this in mind, it is expected that manual and semi-automatic image analysis will gradually be replaced by automatic vision systems as they become more powerful.
The thermal sensor eliminates the disadvantages of changing lighting and requiring active lighting in dark conditions. In addition, in the case of surveillance, the use of thermal imaging does not cause as many privacy issues as the use of visual imaging. However, new types of sensors often present new challenges. For thermal imaging, the lack of texture information may be a disadvantage in some systems, and the reflection of thermal radiation may become a problem in surfaces with high reflectivity. In order for the infrared thermal imager to operate independently in surveillance, an infrared thermal imager with reasonable price, higher resolution, effective optical zoom or wide-angle lens is still needed. To overcome some of these challenges, it may be advantageous to combine thermal images with other image modalities in many applications. However, there is still a lack of a simple, standardized method to calibrate thermal imaging cameras with other sensors. This problem must be solved to make these types of systems practical. Some pre-calibrated thermal imaging camera settings exist today, and it is expected to see more of these combined systems in the future.
As more and more sensors (such as 3D, near infrared and thermal sensors) become available, it is often difficult to choose a vision camera. Thermal sensors have advantages in various applications, and the fusion of different sensors can improve the results in some applications. For the future development of the vision system, careful selection of sensors can open new applications and alternative functions at the same time to improve the performance of current applications.
Post time: Dec-16-2020