Anzeige
Erschienen in:
2019 | OriginalPaper | Buchkapitel
An Image Processing Approach for Analyzing Assessment of Pavement Distress
verfasst von : Surya Pandey, Surekha Dholay
Erschienen in: Innovations in Computer Science and Engineering
Verlag: Springer Singapore
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
Mechanized labeling of pavement distress is of preponderant usefulness in transportation segment for warrant of safety. Typically, non-automated techniques are obligatory for conventional classification algorithms, thus having constrained breadth of usage. In the matter herein presents a modus operandi for finding and classifying pavement distress on road which makes use of a deep neural network technique called as convolutional neural network (CNN) to classify the given images of distress into their different categories by making use of “activation function” to proclaim distinct identification of likely features by selecting the features automatically. A comparative result is given for three activation functions, viz. ReLU (Rectified Linear Unit), Sigmoid, and Tanh. Denouement from the results herein points out that ReLU surpasses Sigmoid and Tanh. Amidst Sigmoid and Tanh, Tanh furnishes exceeding accomplishment in terms of time.