US 12,475,737 B2
Image analysis system for forensic facial comparison
Rubén Martos Fernández, Ponferrada (ES); Enrique Bermejo Nievas, Ponferrada (ES); Oscar Ibáñez Panizo, Ponferrada (ES); Andrea Valsecchi, Ponferrada (ES); Pablo Mesejo Santiago, Granada (ES); Alexsandro Vasconcellos Da Silva, Granada (ES); Fernando Navarro Merino, Granada (ES); María Inmaculada Alemán Aguilera, Granada (ES); Oscar Cordón García, Granada (ES); and Sergio Damas Arroyo, Granada (ES)
Assigned to UNIVERSIDAD DE GRANADA, Granada (ES); and PANACEA COOPERATIVE RESEARCH S. COOP., Ponferrada (ES)
Appl. No. 17/909,268
Filed by UNIVERSIDAD DE GRANADA, Granada (ES); and PANACEA COOPERATIVE RESEARCH S. COOP., Ponferrada (ES)
PCT Filed Mar. 5, 2021, PCT No. PCT/ES2021/070162
§ 371(c)(1), (2) Date Sep. 2, 2022,
PCT Pub. No. WO2021/176126, PCT Pub. Date Sep. 10, 2021.
Claims priority of application No. ES202030191 (ES), filed on Mar. 6, 2020.
Prior Publication US 2023/0088742 A1, Mar. 23, 2023
Int. Cl. G06V 40/16 (2022.01); G06V 10/10 (2022.01); G06V 10/26 (2022.01); G06V 10/34 (2022.01); G06V 10/75 (2022.01); G06V 10/82 (2022.01); G06V 20/70 (2022.01)
CPC G06V 40/172 (2022.01) [G06V 10/16 (2022.01); G06V 10/26 (2022.01); G06V 10/34 (2022.01); G06V 10/82 (2022.01); G06V 20/70 (2022.01)] 9 Claims
OG exemplary drawing
 
1. An image analysis system for forensic facial comparison, comprising:
image acquisition means for acquiring at least one questioned image (I1) of a subject,
processing means with access to the questioned image (I1) and at least one reference image (I2);
wherein the processing means automatically perform the forensic facial comparison by comparing the questioned image (I1) with the reference image (I2), wherein the processing means are configured to perform at least one stage of morphological analysis (I1) of forensic facial comparison by means of neural networks, and a decision-making stage (I5) to obtain a fuzzy value applying soft computing, wherein the fuzzy value is an indicator of the probability that the questioned image (I1) and reference image (I2) correspond to the same subject; and
wherein the morphological analysis step (I1) comprises:
labelling by means of a first neural network the questioned image (I1) and the reference image (I2) with morphological descriptors, said morphological descriptors representing anatomical and/or morphological aspects of a face; comparing the morphological descriptors labelled within the questioned facial image (I1) and within the reference facial image (I2) and calculating a degree of similarity between the morphological descriptors labelled within the questioned facial image (I1) and within the reference facial image (I2), obtaining a value x11;
detecting by means of a second neural network individualizing elements in the questioned image (I1) and in the reference image (I2), comparing the individualizing elements detected in the questioned facial image (I1) and in the reference facial image (I2) and obtaining a degree of similarity between the individualizing elements detected in the questioned facial image (I1) and in the reference facial image (I2), obtaining a value x12;
calculating a degree of asymmetry between the questioned image (I1) and the reference image (I2), obtaining a value x13, by calculating the difference between a degree of symmetry of the questioned image (I1) and a degree of symmetry of the reference image (I2), the degree of symmetry of the questioned image (I1) calculated as the difference between the questioned image (I1) and its mirror image, and the degree of symmetry of the reference image (I2) calculated as the difference between the reference image (I2) and its mirror image;
obtaining a value, E1, which is an overall indicator of the forensic facial comparison stage of morphological analysis (I1), wherein E1=x11*w11+x12*w12+x13 w13, wherein w11, w12 and w13 are configurable weights that meet w12>w11>w13.