Anzeige
Erschienen in:
2020 | OriginalPaper | Buchkapitel
Research on Accuracy of Flower Recognition Application Based on Convolutional Neural Network
verfasst von : Jing-Hua Han, Chen Jin, Li-Sha Wu
Erschienen in: Advances in Artificial Intelligence, Software and Systems Engineering
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
Compared with traditional flower recognition methods, the existing flower recognition applications on the market use advanced deep learning technology to improve the accuracy of plant recognition and solve the problem of plant recognition. The article studied the five applications that users commonly use, comparing and analyzing their recognition accuracy, and finally putting forward the feasibility advice for further improvement of flower recognition applications. The method of sampling survey was adopted, this paper divides the garden flowers and wild flowers into different levels according to their common degrees. Each type of flower was shot from 5 different angles and scenes, and recognized by these five applications separately. The results showed that the rankings of the five applications evaluated were Hua Bangzhu, Hua Banlv, Xing Se, Microsoft’s Flower Recognition, and Find Flower Recognition. At pre-sent, it is necessary to continuously improve from the aspects of technology, products and plant libraries.