1. The complete title of one (or more) paper(s) published in the open literature describing the work that the author claims describes a human-competitive result; Title of the first paper: “Design of estimators for restoration of images degraded by haze using genetic programming”. Title of the second paper: “Real-time image dehazing using genetic programming”. %--------------------------------------------------------------------------------------------------------------------- 2. The name, complete physical mailing address, e-mail address, and phone number of EACH author of EACH paper(s); Jose Enrique Hernandez-Beltran, Ave. Instituto Politécnico Nacional 1310, Tijuana, B.C. 22435, Mexico, jhernandez@citedi.mx, +52 664 623 13 44. Victor H. Diaz-Ramirez, Ave. Instituto Politécnico Nacional 1310, Tijuana, B.C. 22435, Mexico, vhdiaz@citedi.mx, +52 664 623 13 44. Leonardo Trujillo, Blvd. Industrial y Ave. ITR Tijuana S/N, Mesa de Otay, Tijuana, B.C. 22500, Mexico, leonardo.trujillo@tectijuana.edu.mx, + 52 664 338 83 81. Pierrick Legrand, 200, Avenue de la Vieille Tour, 33405, Talence, France, pierrick.legrand@u-bordeaux.fr, +33 666 15 24 89. Rigoberto Juarez-Salazar, Ave. Instituto Politécnico Nacional 1310, Tijuana, B.C. 22435, Mexico, rjuarez@citedi.mx, +52 664 623 13 44. %--------------------------------------------------------------------------------------------------------------------- 3. The name of the corresponding author; Victor H. Diaz-Ramirez. %--------------------------------------------------------------------------------------------------------------------- 4. The abstract of the paper(s); Abstract of the first paper: Restoring hazy images is challenging since it must account for several physical factors that are related to the image formation process. Existing analytical methods can only provide partial solutions because they rely on assumptions that may not be valid in practice. This research presents an effective method for restoring hazy images based on genetic programming. Using basic mathematical operators several computer programs that estimate the medium transmission function of hazy scenes are automatically evolved. Afterwards, image restoration is performed using the estimated transmission function in a physics-based restoration model. The proposed estimators are optimized with respect to the mean-absolute-error. Thus, the effects of haze are effectively removed while minimizing overprocessing artifacts. The performance of the evolved GP estimators given in terms of objective metrics and a subjective visual criterion, is evaluated on synthetic and real-life hazy images. Comparisons are carried out with state-of-the-art methods, showing that the evolved estimators can outperform these methods without incurring a loss in efficiency, and in most scenarios achieving improved performance that is statistically significant. Abstract of the second paper: A real-time system for restoration of images degraded by haze is presented. First, a transmission function estimator is automatically constructed using genetic programming. Next, the resultant estimator is employed to compute the transmission function of the scene by processing an input hazy image. Finally, the estimated transmission function and the hazy image are used in a restoration model based on atmospheric optics to obtain a haze-free image. The proposed method is implemented in a laboratory prototype for high-rate image processing. The performance of the proposed approach is evaluated in terms of objective metrics using synthetic and real-world images. %--------------------------------------------------------------------------------------------------------------------- 5. A list containing one or more of the eight letters (A, B, C, D, E, F, G, or H) that correspond to the criteria that the author claims that the work satisfies; (B) The result is equal to or better than a result that was accepted as a new scientific result at the time when it was published in a peer-reviewed scientific journal. (D) The result is publishable in its own right as a new scientific result — independent of the fact that the result was mechanically created. (F) The result is equal to or better than a result that was considered an achievement in its field at the time it was first discovered. (G) The result solves a problem of indisputable difficulty in its field. %--------------------------------------------------------------------------------------------------------------------- 6. A statement stating why the result satisfies the criteria that the contestant claims; (B) The result is equal to or better than a result that was accepted as a new scientific result at the time when it was published in a peer-reviewed scientific journal. The proposed approach provides an effective solution for the challenging problem of restoration of images degraded by haze. We proposed a methodology based on Genetic Programming (GP) that allows the design of efficient and effective estimators of the optical transmission function of an observed hazy image. The estimated transmission function obtained with the evolved estimators allows to retrieve accurately a haze-free image from an observed hazy image. Due to its high impact and wide applicability, the problem of image dehazing has been intensively investigated in the last years by several researchers in the field of image processing. As a result, different successful methods have been published in many prestigious peer-reviewed scientific journals, for instance: 1) Diaz-Ramirez, V. H., Hernandez-Beltran, J. E., & Juarez-Salazar, R. (2019). Real-time haze removal in monocular images using locally adaptive processing. Journal of Real-Time Image Processing, 16(6), 1959-1973. 2) Li, Z., & Zheng, J. (2015). Edge-preserving decomposition-based single image haze removal. IEEE transactions on image processing, 24(12), 5432-5441. 3) Zhu, Q., Mai, J., & Shao, L. (2015). A fast single image haze removal algorithm using color attenuation prior. IEEE transactions on image processing, 24(11), 3522-3533. 4) Ancuti, C. O., & Ancuti, C. (2013). Single image dehazing by multi-scale fusion. IEEE Transactions on Image Processing, 22(8), 3271-3282. 5) He, K., Sun, J., & Tang, X. (2010). Single image haze removal using dark channel prior. IEEE transactions on pattern analysis and machine intelligence, 33(12), 2341-2353. The proposed estimators outperformed the results obtained by state-of-the-art methods (He et al. 2010 and Zhu et al. 2015) in terms of objective metrics and the subjective visual criterion, without sacrificing computational efficiency. Additionally, since the evolved estimators are suitable for implementation in high-performance embedded processors exploiting massive parallelism, real-time operation speed is achieved. (D) The result is publishable in its own right as a new scientific result — independent of the fact that the result was mechanically created. In the performed research, we posed and solved a classic challenging problem of image processing, called single image dehazing, as a supervised learning problem. In this context, we utilized GP techniques to construct new evolved estimators of the optical transmission function of an observed hazy scene. The estimated transmission function with the evolved estimators is used in an optics-based restoration model to retrieve accurately a haze-free image of the scene. The obtained results showed that the problem of single image dehazing using the proposed GP-based approach can be solved more accurately than state-of-the-art hand-crafted solutions and without sacrificing computational efficiency. Moreover, in the proposed GP-based approach, two problem variants were investigated, first, by posing the problem at the descriptor level and second, by posing the problem at a pixel level. The proposed problem formulation permits researchers around the world to solve similar versions of the problem, and test other machine learning methods or GP-based approaches. Additionally, by choosing properly the training images, we were able to generate solutions that generalize well while using a rather small training set. Moreover, by using a bloat-free method, namely, neat-GP, we produced highly accurate and efficient estimators of the optical transmission function of a degraded scene for image dehazing. By performing exhaustive statistical tests, we verified that the evolved estimators outperformed state-of-the-art handmade methods in terms of objective performance measures. Additionally, by performing an efficient implementation of the proposed GP-based approach in a massive parallelism embedded processor real-time image dehazing was achieved. Furthermore, we show for the first time that the well-known neat-GP approach can be used to solve a difficult real-world problem of image processing, instead of the more traditional regression and classification tasks on which neat-GP has only been applied before. In existing related literature, there are several machine learning approaches aimed to estimate the optical transmission function of a hazy scene: 1) Luan, Z., Shang, Y., Zhou, X., Shao, Z., Guo, G., & Liu, X. (2017). Fast single image dehazing based on a regression model. Neurocomputing, 245, 10-22. 2) Cai, B., Xu, X., Jia, K., Qing, C., & Tao, D. (2016). Dehazenet: An end-to-end system for single image haze removal. IEEE Transactions on Image Processing, 25(11), 5187-5198. 3) Ren, W., Liu, S., Zhang, H., Pan, J., Cao, X., & Yang, M. H. (2016). Single image dehazing via multi-scale convolutional neural networks. In European conference on computer vision (pp. 154-169). Springer, Cham. The main drawback of these neural-based methods is that they require large synthetic datasets in order to train and evaluate their learning models. Also, since these synthesized models cannot be analytically analyzed they can be seen merely as black-box solutions. In the proposed GP approach, we only require a small number of training images to evolve feasible estimators and the obtained solutions can be analytically analyzed. (F) The result is equal to or better than a result that was considered an achievement in its field at the time it was first discovered. The results obtained by the proposed evolved estimators outperformed in terms of objective measures and the subjective visual criterion the results obtained by state-of-the-art methods, namely, He et al. 2010 and Zhu et al. 2015. Both of these methods were considered an achievement in the field of image dehazing at the time they were published. (G) The result solves a problem of indisputable difficulty in its field. The proposed approach provides an effective and efficient solution for the challenging problem of restoration of images degraded by haze. In this restoration problem, an input hazy image is usually represented by a mathematical model based on atmospheric optics. Basically, the problem of single image dehazing consist in obtaining an estimate of a haze-free image by processing a single captured image of a hazy scene. It should be noted that this problem is ill-posed because several physical factors such as the depth distribution of the scene, the concentration density of suspended scattering particles in the medium and the magnitude of environmental light, among others, need to be estimated from a single image of the degraded scene. Note that under these circumstances the design of feasible solutions for this problem is very challenging. To date, the image dehazing problem remains open within the field of image processing and many researchers around the world are proposing new approaches and solutions to this challenging research problem. %--------------------------------------------------------------------------------------------------------------------- 7. A full citation of the paper (that is, author names; publication date; name of journal, conference, technical report, thesis, book, or book chapter; name of editors, if applicable, of the journal or edited book; publisher name; publisher city; page numbers, if applicable); Full citation of the first paper: Hernandez-Beltran, J. E., Diaz-Ramirez, V. H., Trujillo, L., & Legrand, P. (2019). Design of estimators for restoration of images degraded by haze using genetic programming. Swarm and evolutionary computation, 44, 49-63. Full citation of the second paper: Hernandez-Beltran, J. E., Diaz-Ramirez, V. H., & Juarez-Salazar, R. (2019). Real-time image dehazing using genetic programming. In Optics and Photonics for Information Processing XIII (Vol. 11136, p. 111360V). International Society for Optics and Photonics. %--------------------------------------------------------------------------------------------------------------------- 8. A statement either that "any prize money, if any, is to be divided equally among the co-authors" OR a specific percentage breakdown as to how the prize money, if any, is to be divided among the co-authors; Jose Enrique Hernandez-Beltran: 40% Victor H. Diaz-Ramirez: 15% Leonardo Trujillo: 15% Pierrick Legrand: 15% Rigoberto Juarez-Salazar: 15% %--------------------------------------------------------------------------------------------------------------------- 9. A statement stating why the authors expect that their entry would be the "best": We believe that our work could qualify as the best for the following reasons: - Our work solves effectively one of the most challenging problems in the image processing field, widely known as single image dehazing. - The results obtained by the proposed GP approach have been published in a high-impact scientific journal in the field of evolutionary computation and in a leading conference in the field of optics and photonics. - The proposed GP approach can evolve accurate and efficient estimators of the optical transmission function of a hazy scene that outperforms existing state-of-the-art handmade methods in terms of accuracy of image restoration. - The proposed methodology showed for the first time that GP can be effectively employed for the design of feasible estimators of the optical transmission function, to solve a difficult real-world problem of image processing. The proposed methodology opens the possibility to design new high-performance estimators by using different configurations of the proposed GP approach. - The evolved estimators can be applied for real-time image processing in a wide range of trending technologies such as: autonomous driving, low-earth-orbit remote sensing systems, light cargo delivery drones, among others. %--------------------------------------------------------------------------------------------------------------------- 10. An indication of the general type of genetic or evolutionary computation used, such as GA (genetic algorithms), GP (genetic programming), ES (evolution strategies), EP (evolutionary programming), LCS (learning classifier systems), GE (grammatical evolution), GEP (gene expression programming), DE (differential evolution), etc. Genetic programming (GP). %--------------------------------------------------------------------------------------------------------------------- 11. The date of publication of each paper. If the date of publication is not on or before the deadline for submission, but instead, the paper has been unconditionally accepted for publication and is "in press" by the deadline for this competition, the entry must include a copy of the documentation establishing that the paper meets the "in press" requirement. First paper: February 2019. Second paper: September 2019.