Design And Implementation Of Database In Flower Identification System

The purpose of flower recognition is to recognize natural flowers through machine learning and pattern recognition.From the flower recognition in the 1990s to the present,it has a long history of development,but the identification of large-scale natural flowers is still in its infancy.A good flower recognition system needs a complete database as support.At present,the size of the flower recognition database will directly affect the accuracy of flower recognition.Under the premise of ensuring the database is large enough,the integrity of the data storage should also be implemented to ensure that the data will not be lost due to accidents.It should also be considered.Due to the large amount of data,unreasonable table space design caused by the slow query and other issues.Therefore,solving the above problems during the process of designing the database is the key to improving the performance of the flower recognition system.Firstly,the thesis introduces the status quo of domestic and foreign research on flower recognition and database,and the necessity of choosing this topic.It puts forward the goal and overall plan of database design in the flower recognition system,and uses the Real Application Clusters cluster architecture unique to ORACLE database.RAC).Explained the specific flow of the entire flower system database design,according to the strategy,analysis,design,construction,documentation,transplantation and product seven stages to design the database.Combined with the specific needs of the flower recognition system,the above-mentioned seven database design stages are analyzed in detail,in particular,the database is designed in three aspects:conceptual design,logical design,and physical design.Especially in the aspect of physical design,the different features of different indexes are discussed.For the flower recognition system,the table space is reasonably designed and the appropriate database index is selected.By building a flower recognition database in a RAC cluster environment,multiple databases are parallelized,the production cost is reduced,and the high availability and high concurrency of the database are improved.Create a B-tree index in the database and define an index organization table to increase query speed while reducing the use of storage space.This paper finally achieved the improvement of the query efficiency of the flower recognition system,completed the high concurrency of the database,established a complete flower recognition database system,and promoted the continuous development of large-scale natural flower recognition.