Vehicle Object Detection in Remote Sensing Imagery Based on Multi-Perspective Convolutional Neural Network

Abstract

Most traditional object detection approaches have a deficiency of features, slow detection speed, and high false-alarm rate. To solve these problems, we propose a multi-perspective convolutional neural network (Multi-PerNet) to extract remote sensing imagery features. Regions with CNN features (R-CNN) is a milestone in applying CNN method to object detection. With the help of the great feature extraction and classification performance of CNN, the transformation of object detection problem is realized by the Region Proposal method. Multi-PerNet trains a vehicle object detection model in remote sensing imagery based on Faster R-CNN. During model training, sample images and the labels are inputs, and the output is a detection model. First, Multi-PerNet extracts the feature map. Meanwhile, the area distribution and object-area aspect ratio in the sample images are obtained by k-means clustering. Then, the Faster R-CNN region proposal network generates the candidate windows based on the k-means clustering results. Features of candidate windows can be obtained by mapping candidate windows to the feature map. Finally, the candidate window and its features are inputted to the classifier to be trained to obtain the detection model. Experiment results show that the Multi-PerNet model detection accuracy is improved by 10.1% compared with the model obtained by ZF-net and 1.6% compared with the model obtained by PVANet. Moreover, the model size is reduced by 21.3%.

Vehicle detection of multi-source remote sensing data using active fine-tuning network

“…The recent success of CNN-based architecture brings the power in vehicle detection, owing to sufficient well-annotated samples (Yang et al, 2018;Ji et al, 2019;Mandal et al, 2019;Schilling et al, 2018). However, costly manual labeling makes it difficult to acquire a large number of labeled samples in practice, leading to the poor detection performance of the previous network-based methods, e.g., FCN (Schilling et al, 2018).…”

“…Visible (VIS) remote sensing imagery uses electromagnetic wave imaging at a wavelength of 400 ∼ 760 nm, which visually reflects the true color and texture of vehicle objects. Inspired by the recent advancement of deep learning and availability of multi-source remote sensing, such as RGB, hyperspectral (Hong et al, 2019a), multispectral (Weng et al, 2014), and synthetic aperture radar (SAR) (Kang et al, 2020), many state-of-the-art vehicle detection algorithms in remote sensing (Cheng et al, 2018;Yang et al, 2018;Wu et al, 2019;Kang and Gellert, 2015;Audebert et al, 2017;Gintautas et al, 2009;Wei et al, 2013;Cheng et al, 2020;Wu et al, 2020) have been developed.…”

Deep Neural Networks Approach for Monitoring Vehicles on the Highway

“…RELATED WORK Detection of vehicles by category inference on video sequence data is an important but challenging task in an urban traffic surveillance system with the primary goal of extracting vehicle features from videos or images captured by traffic surveillance, then identifying vehicle types, and providing reference information for monitoring and traffic control, some researchers propose a deep learning approach using neural networks which in this decade have made progress in vehicle detection, such as (Yang, Li, & Lin, 2018) proposing a multiperspective convolutional neural network (Multi-PerNet) to extract features Remote visual image of vehicle object detection, the Multi-PerNet Model extracts feature maps, while k-means clustering is used as area distribution and object-area aspect ratio in sample images, Faster-R-CNN framework is applied as object classification and detection models . In the work (H. Wang &Cai, 2014) proposed a deep belief network (DBN) architecture for vehicle classification, (Sang et al, 2018) proposed an increase in YOLOv2 performance in vehicle detection, the k-means ++ grouping algorithm is used to group boxes.…”

Rock Classification from Field Image Patches Analyzed Using a Deep Convolutional Neural Network

“…Deep convolutional neural networks (CNNs) are able to automatically learn the features required for image classification from training-image data, thus improving classification accuracy and efficiency without relying on artificial feature selection. Very recent studies have proposed deep learning algorithms to achieve significant empirical improvements in areas such as image classification [14], object detection [15], human behavior recognition [16,17], speech recognition [18,19], traffic signal recognition [20,21], clinical diagnosis [22,23], and plant disease identification [11,24]. The successes of applying CNNs to image recognition have led geologists to investigate their use in identifying rock types [8,9,25], and deep learning has been used in several studies to identify the rock types from images.…”