| Multiocular restauration of images degraded by optical scattering (2017-2021) |
| Jose Enrique Hernandez Beltran, Doctorate of Sciences in Digital Systems, Instituto Politécnico Nacional. |
| Victor H. Diaz-Ramirez, and Rigoberto Juarez-Salazar (Advisers). |
Palabras clave
Image restoration, multiocular vision, optical scattering,
stereo matching, adaptive correlation filters.
Resumen
This work presents an algorithm for restoration of images degraded by optical scattering using a stereo vision. First, given a pair of rectified stereo images of a degraded scene, the disparity map of the scene is estimated using an adaptive correlation filter. The correlation filter detects and estimates the location of a target object in an observed scene, given a reference image which contains the target object defined implicitly. Afterwards, the estimated disparity and the physical parameters of the stereo vision system are used to estimate the depth distribution object objects located in the scene regarding to the vision system.
The depth information is utilized to estimate the atmospheric degradation parameters of the scene. For this, statistical estimators of the atmospheric parameters of the degradation model are proposed, using assumptions inherent to the capture process of stereo images. Finally, the depth of the scene and the estimated atmospheric parameters are used in a restoration model based on the image formation process, to retrieve a clear image of the observed scene. The performance of the proposed algorithm for restoration of images degraded by optical scattering is evaluated using images from an academic database and images captured in an experimental test platform, in terms of objective metrics.
Contributions
Jose Enrique Hernandez-Beltran, V.H. Diaz-Ramirez, and Rigoberto Juarez-Salazar, “Adaptive matched filter for implicit-target recognition: application in three-dimensional reconstruction,” Applied Optics, vol. 58(32), pp. 8920-8930, 2019.
Jose Enrique Hernandez-Beltran, V.H. Diaz-Ramirez, Leonardo Trujillo, and Pierrick Legrand, “Design of estimators for restoration of images degraded by haze using genetic programming,” Swarm and Evolutionary Computation, vol. 44, pp. 49-63, 2019.
Rigoberto Juarez-Salazar, Ceciibet Mendoza-Rodriguez, Jose E Hernandez-Beltran, and Carlos Robledo-Sanchez, “How do phase-shifting algorithms work?,” European Journal of Physics, vol. 39(6), pp. 065302, 2018.
Victor H. Diaz-Ramirez, Jose Enrique Hernandez-Beltran, and Rigoberto Juarez-Salazar, “Real-time haze removal in monocular images using locally adaptive processing,” Journal of Real-Time Image Processing, vol. 16, pp. 1959-1973, 2019.
Jose Enrique Hernandez-Beltran, Mack O. Lugo-Aguilar, Victor H. Diaz-Ramirez, and Rigoberto Juarez-Salazar, “Stereo vision parallelization for three-dimensional reconstruction,” Proc. SPIE, vol. 11509, pp. 1150906, 2020..
Jose Enrique Hernandez-Beltran, Victor H. Diaz-Ramirez, and Rigoberto Juarez-Salazar, “Real-time image dehazing using genetic programming,” Proc. SPIE, vol. 11136, pp. 111360V, 2019.
Jose Enrique Hernandez-Beltran, Victor H. Diaz-Ramirez, and Rigoberto Juarez-Salazar, “Stereo matching using adaptive windows and correlation filtering,” Proc. SPIE, Vol. 10751, pp. 1075102, 2018.
Jose Enrique Hernandez-Beltran, Victor H. Diaz-Ramirez, and Rigoberto Juarez-Salazar, “Restoration of degraded images using stereo vision,” Proc. SPIE, vol. 10395, pp. 103950H, 2017.