Bao-Jen Pong¹*, Shao-Tang Hung¹, Li-Wei Wen², and Chao-Hua Wen²
¹ Industrial Technology Research Institute, Hsinchu, Taiwan
² Xenosquid Technology Co. Ltd., Miaoli, Taiwan
* Corresponding author: bjpong@itri.org.tw
Keywords: Color characterization, High dynamic range, Micro-imaging system
Poster presentation video: https://youtu.be/cxMU2zaMLGk
New technology has driven displays to incorporate richer color volumes, higher luminance, greater contrast, and more pixels, thereby enabling a wide range of new visual experiences and rising an emerging demand of fast and high-dynamic-range (HDR) color image metrology. This is a growing trend, as the applications for the color measurement systems continue to performance become even greater and faster. In this paper, the color characterization of the HDR color micro-imaging system will be addressed. First, we perform a series of experiments to calibrate the luminance and the color volume of a remote and compact RGBW LED Light as the reference light source. Second, we calculate the magnifications of the zoomable telecentric imaging sensor with the dimension calibration patterns, and correct the position of the XY stage movement. Third, we present a method of recovering high dynamic range radiance maps from photographs taken with the telecentric imaging equipment. In the method, multiple photographs of the device-under-test are taken with different amounts of exposure. The algorithm and the color characterization use those differently exposed photographs to recover the response function of the telecentric imaging processes, up to factor of scale, using the assumption of reciprocity. With the resultant response function, the algorithm can merge the multiple photographs into a single, high dynamic range radiance map whose pixel values are proportional to the true radiance values in the scene. The results revealed that acceptable levels of the color difference range from dim scene to bright scene. Forth, we demonstrate the method on images acquired with both micro-LED source and high-resolution display imaging processes. The paper discusses how the framework is applicable in many areas of computer vision involving image sensor calibration, including image-based color modeling, image compositing, and image processing. Lastly, the paper demonstrates a typical application of having high dynamic range radiance maps, such as synthesizing realistic HDR lighting system and simulating the performance of the micro-LED displays system.
Comments