A research team from King Saud University presented an innovative scientific study in Frontiers in Plant Science journal – one of the prestigious scientific journals in the field of plant sciences – where they developed a new intelligent system that utilizes artificial intelligence to identify palm tree species through their images without the need for images of their fruits.
In Saudi Arabia, researchers relied on a new system they named DPXception, which can differentiate between different palm tree varieties simply by using images of the palm trees without the need for using fruit images or any other information.
This research is the first of its kind worldwide, as previous methods for identifying palm tree species relied on genetics or fruits, which entail high costs, specialized expertise, and are not compatible with fruit availability throughout the year.
The intelligent DPXception system analyzes images of palm trees taken using deep learning networks, a technique within artificial intelligence, to extract distinctive features of each type of palm tree. The system was trained on a wide range of images taken from original palm trees, enabling it to accurately identify palm tree species even in the absence of fruits.
The researchers also developed an Android application that utilizes the intelligent DPXception system, allowing for the rapid classification of palm tree species from images captured by the smartphone camera instantly.
Dr. Majdul Sultan bin Safaran, the head of the research team, stated to the Saudi Press Agency (SPA) that the research marks the beginning of a series of research and technical systems that the team is working on applying to palm trees to identify species, threats, and other important aspects.
He pointed out that the team had previously published a scientific study on using artificial intelligence to identify camel breeds through their images as well, and also affirmed that camels and palm trees are symbols of the culture and identity of the nation. Hence, the team is interested in developing scientific and technological research in these two fields.
Dr. Sultan Al-Farhood, who participated in the research, emphasized that this study will contribute to enhancing scientific research in the field of palm trees at both the local and global levels. He indicated that the research team is working on developing other related research paths, such as automatically determining palm tree characteristics and discovering their pests, which will positively impact the quality of date production in Saudi Arabia.
He also noted that the team is also busy developing smart applications based on this system to provide services and advice to farmers, investors, and individuals interested in palm tree cultivation in and outside the Kingdom.
