NETD (Noise Equivalent Temperature Difference) is a key parameter for evaluating MWIR and LWIR infrared cameras. It is a value that represents the signal-to-noise ratio of the temperature difference, which is equivalent to the instantaneous noise of the camera. Therefore, it is approximately the smallest temperature difference that the camera can resolve. It is directly related to the clarity of the camera's measurement. The smaller the value of NETD, the higher its sensitivity and the clearer the image.

NEDT (Noise Equivalent Temperature Difference) is a key parameter for evaluating MWIR and LWIR infrared cameras. It is a value that represents the signal-to-noise ratio of the temperature difference, which is equivalent to the instantaneous noise of the camera. Therefore, it is approximately the smallest temperature difference that the camera can resolve.

It is the instantaneous noise divided by the responsivity, usually expressed in mK. This value is a function of the camera's aperture, integration time and the specific temperature at the time of measurement. To measure it you need a high-quality surface source blackbody. Set the blackbody to 25°C. The camera should be placed at the closest distance (2 to 5 cm). For mid-IR cameras, set the integration time so that the dynamic range of the camera detector (FPA) is about halfway, and do not install the lens.

For uncooled long-wave IR cameras, set the integration time to the maximum value and install a standard 25mm lens. A 2-point non-uniformity correction will be performed to obtain a uniform image. And 3 sets of data will be obtained through the digital output of the camera.

For the first set of data, the black body temperature is set to 20°C and 64 frames of data are collected continuously. The average value of these frames is calculated to obtain a new array composed of each pixel, which represents the response value at 20°C. Its unit is A/D count value.

For the second set of data, the black body temperature is set to 25°C and 128 frames of data are collected continuously. The standard deviation of the data of each pixel in the 128 frames is taken. Then these standard deviation values are used as a matrix to represent the instantaneous noise. Its unit is A/D count value.

For the third set of data, the blackbody temperature was set to 30°C and 64 frames of data were collected continuously as in the first set of data. The average of these frames was calculated to obtain a new array of each pixel, which represents the 30°C response value. Its unit is A/D count value. Subtract the 20°C response matrix from the 30°C response matrix and divide by 10 to obtain a responsivity matrix with units of counts per degree. The instantaneous noise matrix was then divided by the responsivity matrix to obtain the NEDT matrix with units of K. After calculating the average of all pixels (eliminating bad pixels), multiply by 1000 to obtain the value in mK.