Facial biometrics play an essential role in the fields of law enforcement and forensic sciences. When comparing facial traits for human identification in photographs or videos, the analysis must account for several factors that impair the application of common identification techniques, such as illumination, pose, or expression. In particular, facial attributes can drastically change depending on the distance between the subject and the camera at the time of the picture. This effect is known as perspective distortion, which can severely affect the outcome of the comparative analysis. Hence, knowing the subject-to-camera distance of the original scene where the photograph was taken can help determine the degree of distortion, improve the accuracy of computer-aided recognition tools, and increase the reliability of human identification and further analyses. In this paper, we propose a deep learning approach to estimate the subject-to-camera distance of facial photographs: FacialSCDnet. Furthermore, we introduce a novel evaluation metric designed to guide the learning process, based on changes in facial distortion at different distances. To validate our proposal, we collected a novel dataset of facial photographs taken at several distances using both synthetic and real data. Our approach is fully automatic and can provide a numerical distance estimation for up to six meters, beyond which changes in facial distortion are not significant. The proposed method achieves an accurate estimation, with an average error below 6 cm of subject-to-camera distance for facial photographs in any frontal or lateral head pose, robust to facial hair, glasses, and partial occlusion. • Accurate estimation of subject-to-camera distance in portrait photographs. • A novel metric is proposed, based on the effects of perspective in facial distortion. • A new database of facial images at a distance is introduced for human identification. • A transfer learning approach overcomes the limitations of current methods. • Robust to expression, occlusion and pose without requiring anatomical information.

Comparative radiography is the forensic anthropology technique in which ante-mortem (AM) and post-mortem (PM) radiographic materials (e.g., X-ray images or CTs) are compared in order to determine the identity of a deceased human being. One of the most commonly used anatomical structures in comparative radiography are the frontal sinuses. The frontal sinuses are osseous cavities located in the skull, which are used in forensic identification tasks due to their singularity and high identification power. In order to automate the comparison of frontal sinuses in AM and PM materials, it is necessary to perform the registration of these materials (i.e., it is necessary to carry out the alignment of these anatomical regions). However, the manual alignment of these structures is a time-consuming and subjective process. In order to tackle this problem, this paper presents an automatic frontal sinuses registration method in comparative radiography using real-coded evolutionary algorithms (RCEAs). The task is formulated as a 2D-3D image registration problem using a 9 Degrees of Freedom perspective transformation model; two RCEAs (DE and MVMOSH) are compared in the minimization of the registration cost function, and the best of them (MVMO-SH) is applied to an identification scenario including 50X-ray images and 50 CTs. The results obtained show that the proposed automatic identification system is able to filter more than 80% of the sample.

The study of the physical properties of porous media is crucial for petrophysics laboratories. Even though micro computed tomography (CT) could be useful, the appropriate evaluation of flow properties would involve the acquisition of a large number of representative images. That is often unfeasible. Stochastic reconstruction methods aim to generate novel, realistic rock images from a small sample, thus avoiding a large acquisition process. In this contribution, we improve a previous method for 3D-to-3D reconstruction of the structure of porous media by applying generative adversarial neural networks (GANs). We compare several measures of pore morphology between simulated and acquired images. Experiments include Beadpack, Berea sandstone, and Ketton limestone images. Results show that our GANs-based method can reconstruct three-dimensional images of porous media at different scales that are representative of the morphology of the original images. Furthermore, the generation of multiple images is much faster than classical image reconstruction methods.

Craniofacial superimposition aims to identify a missing person by comparing its skull with photos of possible candidates. Among the difficult tasks involved, this requires superimposing the skull over each photo, matching the pose of the skull with that of the face, a problem known as skull-face overlay (SFO). Several computerized methods for SFO have been proposed, in an effort to relieve forensic experts of this complex, time-consuming, and subjective task. A system to generate artificial SFO data from computed tomography images has been also introduced, providing researchers with data to test and compare their techniques more reliably. This paper improves the state of the art on both fronts. We introduce a novel SFO algorithm that is substantially more accurate, more reliable, and much faster than the existing methods. An extensive experimental study and statistical analysis validates our findings. Moreover, we propose an improved method to simulate SFO data which, by replacing real photos with simulated ones, is able to generate a wider range of scenarios. This module provides complete control over the pose of the subject and the camera parameters, and even the ability to reproduce inter-expert errors in processing the input data, leading to a more controlled and thorough testing under realistic conditions.

Image registration (IR) or image alignment is a fundamental step in medical image analysis when multiple images are involved. In most of such applications, the registration is performed following the intensity-based approach, which turns IR into a complex, computationally expensive, continuous optimization problem. In this paper, we introduce a new technique for intensity-based medical IR using the Bacterial Foraging Optimization Algorithm (BFOA), a novel bio-inspired metaheuristic. BFOA has recently obtained promising results in many real-world applications, including feature-based IR. The new algorithm is compared on a complex medical IR application against recent, outstanding IR techniques both traditional and based on meta-heuristics. The results show that our proposal is competitive with the state of the art, making BFOA a promising solution to tackle other complex, real-world optimization problems.

This paper describes a hybrid level set approach to medical image segmentation. The method combines region-and edge-based information with the prior shape knowledge introduced using deformable registration. A parameter tuning mechanism, based on Genetic Algorithms, provides the ability to automatically adapt the level set to different segmentation tasks. Provided with a set of examples, the GA learns the correct weights for each image feature used in the segmentation. The algorithm has been tested over four different medical datasets across three image modalities. Our approach has shown significantly more accurate results in comparison with six state-of-the-art segmentation methods. The contributions of both the image registration and the parameter learning steps to the overall performance of the method have also been analyzed.

Image registration is a fundamental step in combining information from multiple images in medical imaging, computer vision and image processing. In this paper, we configure a recent evolutionary algorithm for medical image registration, r-GA, with an offline automatic parameter tuning technique. In addition, we demonstrate the use of automatic tuning to compare different registration algorithms, since it allows to consider results that are not affected by the ability and efforts invested by the designers in configuring the different algorithms, a crucial task that strongly impacts their performance. Our experimental study is carried out on a large dataset of brain MRI, on which we compare the performance of r-GA with four classic IR techniques. Our results show that all algorithms benefit from the automatic tuning process and indicate that r-GA performs significantly better than the competitors.

We introduce a new intensity-based image registration (IR) technique based on a modern, real-coded genetic algorithm. Our proposal is tested on 16 registration scenarios involving real-world MRI medical images. A novel methodological framework to compare heterogeneous IR algorithms is also described. Following such methodology, our algorithm is compared with four well-known IR techniques of different natures. The proposed method is able to improve the results of these techniques in the majority of the scenarios.

The Sharpened No Free Lunch theorem states that all optimization algorithms have the same performance on sets of functions that are closed under permutation, independently of the considered performance measure. However, not all performance measures are informative on how fast or how accurately an algorithm can solve a given problem. In this paper we focus on a particular performance measure, called optimization speed, that quantifies how fast a search algorithm is able to find an optimal solution, and we try to characterize the set of functions on which all possible search algorithms have the same optimization speed. We call fair these sets, and we prove some results about their structure, the number of such sets and the computational complexity of checking fairness.

Fitness-Proportional Negative Slope Coefficient is a fitness landscapes measure that has recently been introduced as a potential indicator of problem hardness for optimisation. It is inspired to an older measure, the Negative Slope Coefficient, and it has been theoretically modelled. Preliminary experiments have suggested that it may be a good predictor of problem hardness. However, this measure has not undergone any convincing and comprehensive empirical testing. Our objective is to fill this gap. So, we perform empirical tests using a large set of invertible functions of unitation. We find that while this measure may correctly predict the degree of evolvability of a landscape, this does not necessarily correlate with the difficulty of problems. Some landscapes may show, for example, limited evolvability and yet be easy to solve because either solutions are already present in the initial population or the computational resources provided exceed evolvability obstacles. Or it may be impossible to solve them irrespective of their evolvability simply because they are far too vast for the computational resources provided. These situations are hardly captured by the Fitness-Proportional Negative Slope Coefficient.

Cooperative coevolution has proven to be a promising technique for solving complex combinatorial optimization problems. In this paper, we present four different strategies which involve cooperative coevolution of a genetic program and of a population of constants evolved by a genetic algorithm. The genetic program evolves expressions that solve a problem, while the genetic algorithm provides "good" values for the numeric terminal symbols used by those expressions. Experiments have been performed on three symbolic regression problems and on a "real-world" biomedical application. Results are encouraging and confirm that our coevolutionary algorithms can be used effectively in different domains.