Fig. 6: Ultra-high-speed imaging. | Nature Communications

Fig. 6: Ultra-high-speed imaging.

From: All-fiber high-speed image detection enabled by deep learning

Fig. 6

a Waveforms collected during the transient time when the DMD refreshes from an image 3 to an image 0. Insets (a1–a17) show the images reconstructed from some of these waveforms. b Waveforms recorded with the DMD periodically displaying two images, from which we see that each image has an exposure time of 235 μs, which is used by the DMD to load the data of the next image into the memory cells beneath the pixels. After all data have been loaded, the DMD switches to the other image within only 3 μs, as marked by the dashed circle. The details of this region are shown in (a). c Real images of the DMD captured during the refresh process using a commercial high-speed camera. d Structure of one pixel of the DMD. A DMD chip has many micromirrors on its surface, which correspond to the pixels in the image to be displayed. These micromirrors can be individually rotated by approximately ±12° to an on or off state. In the on state, the illumination light is reflected back to the fiber probe. In the off state, the light is directed in another direction. Thus, the array of micromirrors can produce intensity modulations on the light field. When the DMD refreshes its current image, all micromirrors move to their assigned states at the same moment. e Schematic of the patterns of digits 3 and 0 displayed on the DMD chip. When the DMD switches from 3 to 0, the micromirrors in the overlapping region of the two patterns (region ii) do not change their states, while those in regions (i) and (iii) rotate in the opposite directions.

Back to article page