 |
| Infrared Camera |
When seeking to select an infrared camera, is very important to better understand the attributes of this camera is that most impact the quality of the resulting infrared image. This paper covers three main areas that affect the quality of thermal images: pixel resolution, thermal sensitivity and fixed pattern noise. Each region has a significant impact on thermal image quality.
If you buy a digital camera in the past, the purchase you are likely to be influenced by your belief that the number of pixels is the most important specification when trying to assess the image quality between all the options offered by the camera. For anyone who read Consumer Reports ™ and their detailed evaluation of a digital camera you will appreciate that the performance of the camera includes a careful analysis of more than the number of pixels. Because the thermal camera is basically an image converter (heat energy radiation in the visible image), you need to understand what are the main attributes that determine the thermal image quality and how they each contribute to the quality of images that you may experience in your application.
Thermal Sensitivity
NETD is the temperature difference of the same scene with both the internal sound of the detector (detector NETD) or the total electronic noise of a measurement system (system NETD). As a buyer you need to evaluate camera system NETD. Test Setup a black reference temperature control and some kind of ambient (passive) object that creates a simple gap target for the camera to visualize. Black body temperature is adjusted to nearly equal to the target ambient temperature. An oscilloscope measures the analog video output and a horizontal line at the point where the delta between the reference and the ambient temperature is no longer the target creating a NETD measured signal is set by the measured temperature difference between the reference and target ambient reference.
MRTD - Minimum Temperature Difference decided
This is a test system. An observer was asked to assess the difference in minimum temperature at which the target bar 4 can be solved by watching the video output is displayed as a reference point and the target set temperature ambient brought close together. Minimum difference will change with spatial frequency of the target bar is used. A curve obtained MRTD against spatial frequencies that characterizes the performance of imaging systems. Modern infrared imaging systems can have low frequency spatial MRTDs tens of milli-Kelvin.
Benefit from our significantly large format camera you combine the need for high sensitivity at the sight of high spatial frequencies.
To simplify the basics of heat sensitivity let us focus on a single pixel of an infrared sensor in an uncooled infrared camera. Each pixel in a sensor array of uncooled focal plane image is basically a resistor made using MEMS (micro electro mechanical systems).
The basic structure pixel uncooled thermal camera is the microscopic structure of the bridge resistor material and a thin absorbing layer has been saved. Foot bridge deck suspended above an integrated circuit and provide electrical connections between the resistive bridge circuits and silicon readings. Reading of the bias voltage control IC thin-film resistors and multiplexes all the pixel signals out into the electronic imaging camera.
As infrared radiation is absorbed by each pixel changes in temperature as the photon energy (wavelength 8-14 microns) is converted into heat which in turn changes the resistance of thin film resistors that pixel. IC reading sent the stress in every element of "micro-bolometer 'and the signal is proportional to the heat absorbed by each detector is the basis of real time video images.
Electrical circuits of infrared sensors are very simple, the voltage is turned on for each pixel and the change in resistance of thin film resistors based on the temperature of the sample pixel and converted into digital values. All analog signals to bring some level of noise as the signal generated by the sensor. Signal to noise ratio greatly affect the quality of images from the camera because the noise level is usually a fixed amount and the detector with the gain increased the system will begin displaying the voice signal and you'll start seeing "snow" in the picture.
Noise signal level is generally defined as the Noise Equivalent Temperature Difference.
Like the electrical circuit there are many opportunities for electrical noise to enter the system, but the quality (signal to noise) from signals that come directly from the infrared pixel having the greatest influence on thermal sensitivity, because almost all developers have access to a camera for the same electronic components which to create the camera. Therefore, thermal sensitivity is largely based on the quality of the infrared imager array.
Other issues such as the lens f number also impact the thermal sensitivity. Your camera lens is likely ƒ1.0 (same focal length lens diameter) is considered a "fast" lens. By comparison the number f in digital cameras you are likely between ƒ3 and ƒ5 while the camera is used in mobile and low cost system can be as high as ƒ20! As application demands lead to longer focal length lens was practical to go to the "slow" optics to reduce size, weight and cost of a telephoto lens and trade off some thermal sensitivity. For example, an optical f1.4 will decrease the thermal sensitivity of 2X and 4X optical F2.0 reduction in thermal sensitivity. Therefore, systems with 50mK sensitivity using a standard lens will maintain good sensitivity (100mK) when a telephoto lens attached to the camera ƒ1.4 other verses sensitivity thermal camera that starts at 200mK 100mK and a time to see through the "slower" ( f higher than 1 point).
Other issues such as the lens f number also impact the thermal sensitivity. Your camera lens is likely ƒ1.0 (same focal length lens diameter) is considered a "fast" lens. By comparison the number f in digital cameras you are likely between ƒ3 and ƒ5 while the camera is used in mobile and low cost system can be as high as ƒ20! As application demands lead to longer focal length lens was practical to go to the "slow" optics to reduce size, weight and cost of a telephoto lens and trade off some thermal sensitivity. For example, an optical f1.4 will decrease the thermal sensitivity of 2X and 4X optical F2.0 reduction in thermal sensitivity. Therefore, systems with 50mK sensitivity using a standard lens will maintain good sensitivity (100mK) when a telephoto lens attached to the camera ƒ1.4 other verses sensitivity thermal camera that starts at 200mK 100mK and a time to see through the "slower" ( f higher than 1 point).
As you can see from the various issues raised in this paper is very complex nature of thermal sensitivity, but in the real world of the human eye is very good at distinguishing small differences in image quality that you would know it (good sensitivity) when you see it.
Non-Uniformity Correction
Because of the increasing number of pixels and their sensitivity to improve image quality are increasingly dependent on a process called Non Uniformity Calibration or Nuc. As mentioned before we are basically an array of microbolometer imaging arrays of small resistors, and because the micro-scale devices, there are variations in how each pixel to respond to infrared energy from the object. During manufacturing infrared camera sensor must be normalized, which means that the different responses and DC output for each detector should focus out. Thermal camera feature is usually an internal flag or iris which regularly positioned in front of the detector as a reference zero constant temperature differences between the pixels. This is a refinement of the process Nuc plant and is sometimes referred to as "touch-up."
Because touch up the source was in the lens, the extra picture quality as possible when doing touch up the calibration by using either the lens or the camera lens cover to expose a large uniform surface. As the camera performance improve non-uniformity created by the lens will begin to look and to end a simple picture quality through the lens calibration step will ensure the highest image quality camera capable of producing.
High image quality benefits of increased flexibility
• Many of the larger to examine a variety of target distance
• The ability to visualize low-contrast target thermal
• More intuitive diagnosis of heat-related problems
• Improved quality infrared images seen together since a better matching resolution infrared and visible cameras ..
• Flexibility to include lower costs and lighter weight lenses optional
• Diagnosis is more intuitive temperature anomaly
|