2026-04-13T01:29:06Zhttps://recercat.cat/oai/request

oai:recercat.cat:2072/4031322026-04-08T02:33:19Zcom_2072_98col_2072_378192

00925njm 22002777a 4500 dc González-Castro, Víctor author 2014 Advisors: Enrique Alegre and Rocío Alaiz-Rodríguez. Date and location of PhD thesis defense: 16 December 2013, Universidad de León Semen quality assessment is a crucial task in artificial insemination processes, both for humans and animals. Animal artificial insemination allows farmers to save time and money (e.g. working with a limited number of animals). They purchase semen samples to companies devoted to their production and commercialization, but they need these samples to be optimal for fertilization. As a result, semen production centers have to carry out rigorous quality control procedures to guarantee good standards A semen sample with a high proportion of (i) dead spermatozoa, or (ii) sperm heads with damaged acrosomes will have low fertilization potential. Therefore, sperm vitality and acrosome integrity are two of the parameters evaluated by veterinaries in semen quality control processes. Currently, both are assessed manually, by means of a visual process which entails expensive equipment, such as stains and fluorescence microscopes. Moreover, they may be a source of errors, as any manual process is. The goal is to develop an automatic system to estimate the proportions of damaged acrosomes and dead spermatozoa using just a computer and a digital camera connected to a phase contrast microscope, which is affordable by any laboratory. The contributions made on this PhD thesis in the fields of Image Processing and Machine Learning [1] can be helpful for this goal. Concretely, several texture description approaches have been evaluated for this task. Furthermore, a new intelligent segmentation process, an adaptive texture description method, and two robust approaches for estimating class proportions of unlabeled datasets have been proposed. All these methods are applied to automatic boar semen quality estimation. Features and image descriptors Statistical pattern recognition Machine learning and data mining Medical image analysis Applications Adaptive texture description and estimation of the class prior probabilities for seminal quality control