Camera-Based Blind Spot Detection with a General Purpose Lightweight Neural Network

Abstract

Blind spot detection is an important feature of Advanced Driver Assistance Systems (ADAS). In this paper, we provide a camera-based deep learning method that accurately detects other vehicles in the blind spot, replacing the traditional higher cost solution using radars. The recent breakthrough of deep learning algorithms shows extraordinary performance when applied to many computer vision tasks. Many new convolutional neural network (CNN) structures have been proposed and most of the networks are very deep in order to achieve the state-of-art performance when evaluated with benchmarks. However, blind spot detection, as a real-time embedded system application, requires high speed processing and low computational complexity. Hereby, we propose a novel method that transfers blind spot detection to an image classification task. Subsequently, a series of experiments are conducted to design an efficient neural network by comparing some of the latest deep learning models. Furthermore, we create a dataset with more than 10,000 labeled images using the blind spot view camera mounted on a test vehicle. Finally, we train the proposed deep learning model and evaluate its performance on the dataset.

Machine Learning and Embedded Computing in Advanced Driver Assistance Systems (ADAS)

“…Camera-based methods will be much more cost-effective than using a dedicated radar in this application. In addition, a dataset with more than 10,000 labeled images was generated using a blind spot view camera mounted on a test vehicle [9].…”

“…Furthermore, a faster region-based convolutional neural networks (R-CNN) was evaluated using this dataset and a new R-CNN plus tracking technique to accelerate the process of real-time urban object detection was developed and evaluated [15]. A blind spot detection dataset is introduced in [9]. Refer to Section 2.2 for more information, as this paper belongs in both categories.…”

“…Some papers fit into multiple categories (e.g., [9,10]). It is also worth noting that three papers were selected as feature papers for the Special Issue:…”

A Study on Development of the Camera-Based Blind Spot Detection System Using the Deep Learning Methodology

“…In [13], the camera-based blind spot detection system in conjunction with a network based on AlexNet was proposed as a real-time embedded system application. In this network structure, there are four blocks after the first convolutional layer, and each block is designed by various combinations of the concepts based on the Visual Geometry Group (VGG), depthwise separable convolutions, residual learning, and the squeeze-and-excitation module.…”

Vehicular Systems Technologies: Challenges and Trends Across Transportation Means

“…For instance, energy management is another niche of interest in order to enable efficient battery-powered vehicles such as drones, Unmanned Maritime Vehicles (UMV), or electric road-vehicles, among others [5]. Of course, the enhancement of on-board electronic systems for monitoring the status of the vehicles or the transportation infrastructure is also a field in continuous evolution [6,7].…”