Research and Application of Pig Iris Recognition Technology

This paper mainly studies the collection, processing, and recognition of individual iris images of pigs, as well as the implementation of an application management system. The collection, processing, and recognition of iris images include iris image acquisition, iris detection and localization, iris normalization, iris feature extraction, and iris matching. The acquisition of iris images directly affects the subsequent feature extraction and recognition results, and this step is easily affected by factors such as lighting, shooting angle, and eyelid occlusion in many cases. The process of quickly detecting the presence of the iris and locating its position from an iris image is called iris detection and localization. The detected iris texture area needs to be normalized to a fixed rectangular area to reduce interference from factors such as rotation, scaling, and displacement of the iris. Iris feature extraction extracts recognizable and effective information from normalized iris regions, and constructs iris texture feature encoding. The process of identifying by matching iris feature codes with iris codes in the database is called iris matching. The management system functionality and database design module includes individual pig information management, supplier management, and personnel management for pig farms. Database design includes the logical design of the database, data dictionary, and database implementation, among which the logical design of the database includes ER diagram and relational schema. The data dictionary contains data items, data storage entries, and data flow graphs. The database implementation includes the definition of data tables and stored procedures. According to the design requirements of the system, a pig farm management system based on iris recognition technology is developed using Visual Studio 2020 as the development environment and SQL Server 2018 as the database platform.