-
Categories
-
Pharmaceutical Intermediates
-
Active Pharmaceutical Ingredients
-
Food Additives
- Industrial Coatings
- Agrochemicals
- Dyes and Pigments
- Surfactant
- Flavors and Fragrances
- Chemical Reagents
- Catalyst and Auxiliary
- Natural Products
- Inorganic Chemistry
-
Organic Chemistry
-
Biochemical Engineering
- Analytical Chemistry
-
Cosmetic Ingredient
- Water Treatment Chemical
-
Pharmaceutical Intermediates
Promotion
ECHEMI Mall
Wholesale
Weekly Price
Exhibition
News
-
Trade Service
Atmospheric turbulence effects are widely present in daily life and industrial production
.
Under the influence of turbulence, the refractive index of the atmosphere changes irregularly with time and space, resulting in light intensity, phase and directional fluctuations during transmission.
In response to this problem, Professor Bai Xiangzhi’s team conducted semi-physical simulations based on the imaging degradation mechanism of atmospheric turbulence effects, and proposed a reverse reasoning method based on generative countermeasure networks (TSR-WGAN), which suppressed the blurring effect caused by light fluctuations (Figure 2( a)), alleviating the continuous geometric distortion and jitter in the time domain caused by the change of the optical path (Figure 2(b))
.
The PSNR, RRED and other objective quantitative indicators are superior to the comparison method, and the overall score index leads the comparison method by more than 50% in the subjective test involving 50 participants (Figure 3)
Figure 1 Schematic diagram of the physical process of image degradation under the effect of atmospheric turbulence
Figure 2 (a) TSR-WGAN improves the effect of image blurring
.
The left half of the figure is the degraded image affected by atmospheric turbulence, and the right half is the corrected effect of TSR-WGAN (b) TSR-WGAN (the last column) and other comparison methods to correct jitter and geometric distortion
Figure 3 The subjective imaging quality normalized evaluation index results calculated based on the Bradley-Terry model, TSR-WGAN is significantly higher than the comparison method
