Identification of river hydromorphological features using histograms of oriented gradients cascaded to the Viola-Jones algorithm

• Main Authors: Cuevas, Jerome, Chua, Alvin, Sybingco, Edwin, Bakar, Elmi Abu
• Format: text
• Published: Animo Repository 2019
• Subjects: Drone aircraft in remote sensing; Support vector machines; Rivers; Mechanical Engineering
• Online Access: https://animorepository.dlsu.edu.ph/faculty_research/3441; https://animorepository.dlsu.edu.ph/context/faculty_research/article/4443/type/native/viewcontent/ijmerr.8.2.289_292
• Institution: De La Salle University

In this paper, a quadcopter equipped with a camera was used to capture images from a river. These captured images were used as training data in the automated detection program used to identify the hydromorphological features in the area of the river such as trees, roofs, roads and the shore. The histogram of oriented gradient with support vector machine classifier was cascaded with the Viola Jones Algorithm in order to recognize hydromorphological features. Testing was done using different images to verify the effectiveness of the detection system compared with previous studies. System evaluation and success of the cascaded system was determined using the percentage of correct detected features in the image. The results showed that the cascaded system has increased the accuracy compared to the implementation with only the Viola Jones Algorithm. © 2019 Int. J. Mech. Eng. Rob. Res.