An Improved Deep Convolutional Neural Network-Based Autonomous Road Inspection Scheme Using Unmanned Aerial Vehicles

Abstract

Recent advancements in the field of machine learning (ML) provide opportunity to conduct research on autonomous devices for a variety of applications. Intelligent decision-making is a critical task for self-driving systems. An attempt is made in this study to use a deep learning (DL) approach for the early detection of road cracks, potholes, and the yellow lane. The accuracy is not sufficient after training with the default model. To enhance accuracy, a convolutional neural network (CNN) model with 13 convolutional layers, a softmax layer as an output layer, and two fully connected layers (FCN) are constructed. In order to achieve the deeper propagation and to prevent saturation in the training phase, mish activation is employed in the first 12 layers with a rectified linear unit (ReLU) activation function. The upgraded CNN model performs better than the default CNN model in terms of accuracy. For the varied situation, a revised and enriched dataset for road cracks, potholes, and the yellow lane is created. The yellow lane is detected and tracked in order to move the unmanned aerial vehicle (UAV) autonomously by following yellow lane. After identifying a yellow lane, the UAV performs autonomous navigation while concurrently detecting road cracks and potholes using the robot operating system within the UAV. The performance model is benchmarked using performance measures, such as accuracy, sensitivity, F1-score, F2-score, and dice-coefficient, which demonstrate that the suggested technique produces better outcomes.

Technique of Feature Extraction from an image in Face Recognition with Implementation using own created dataset

“…Each cell is in charge of calculating the Bounding box "B." The bounding box comprises five fundamental characteristics: x, y in the middle, width, and height, denoted as h and w, however, and cs, which stands for confidence [57].…”

Efficient Lane Detection Technique Based on Lightweight Attention Deep Neural Network

“…is work can detect lane lines without number limits. In [13], the authors improve you only look once (YOLO) object detector and realize the detection of yellow lane lines by means of object detection. Pan et al [8] proposed a spatial CNN that aggregates the features of each pixel through slice-by-slice convolution in a layer, resulting in top-1 performance in the CVPR'17 TuSimple benchmark.…”

Adaptive Convolutional Neural Networks for Enhanced Memory Retention and Restoration in Optoelectronic Vision Devices

“…An 80% of human sensory data comes from the vision system which presents a powerful tool to guide new hardware development for applications ranging from healthcare, space, and defense to environmental monitoring technologies. [1][2][3][4][5][6] The recent deployment of artificial vision systems in autonomous vehicles and robots has become an important component of intelligent technologies capable of autonomous decision-making. [7] For instance, machine vision systems are now capable of accurately recognizing air writing and hand gestures using machine learning with various architectures such as convolutional neural network (CNN), recurrent neural network, deep belief network, and deep coding networks.…”