Journal of Semiconductors, Volume. 45, Issue 12, 120401(2024)
Artificial cat eye camera for objects detection against complex backgrounds and varied lighting
Fig. 1. (Color online) (a) Schematics of feline and conventional vision during the daytime and (b) nighttime. (c) Schematics of the feline vision in the daytime with a VP for light adaptation. The yellow plane represents the tangential plane, and the blue plane represents the sagittal plane. (d) Optical simulation of the cross-sectional focal spot according to object distance. The horizontal cross section represents the tangential plane, and (e) the vertical cross section represents the sagittal plane. (f) Ray tracing simulation for various object distances with small circular pupils and (g) VPs, the cross-shaped object is located at 150, 200, and 250 mm. (h) Simulation for camouflage breaking with small CPs and VPs. The center of the random texture is located at 200 mm, and the background random texture is located at 400 mm. (i) Photograph of the fabricated hemispherical silicon photodetector array combined with patterned silver reflectors (HPA-AgR). The inset shows an individual photodiode pixel with a circuit diagram. (j) Exploded structure of the device with a detailed thickness of each component. (k) Schematic illustration showing the artificial feline eye–inspired vision system. (l) The tested image with the ground truth (GT) image and noisy GT of letters (i.e., F, O, C, U, and S) obtained with a small CP and VP. (m) Optical simulation results for the dataset with small CP and VP for each label. (n) The calculated accuracy rates for the Fashion-MNIST dataset from image simulations, both with and without noise. Copyright 2024, American Association for the Advancement of Science[10].
Get Citation
Copy Citation Text
Shengqiang Zhang, Zhuoran Wang. Artificial cat eye camera for objects detection against complex backgrounds and varied lighting[J]. Journal of Semiconductors, 2024, 45(12): 120401
Category: Research Articles
Received: Sep. 28, 2024
Accepted: --
Published Online: Jan. 15, 2025
The Author Email: Wang Zhuoran (ZRWang)