scispace - formally typeset
Search or ask a question
Journal Article•DOI•

An Efficient Method for Traffic Sign Recognition Based on Extreme Learning Machine

Zhiyong Huang1, Yuanlong Yu1, Jason Gu2, Huaping Liu3•
Fuzhou University1, Dalhousie University2, Tsinghua University3
01 Apr 2017-IEEE Transactions on Systems, Man, and Cybernetics (IEEE Trans Cybern)-Vol. 47, Iss: 4, pp 920-933
TL;DR: Experimental results have shown that this proposed method obtains not only high recognition accuracy but also extremely high computational efficiency in both training and recognition processes in these three datasets.
Abstract: This paper proposes a computationally efficient method for traffic sign recognition (TSR). This proposed method consists of two modules: 1) extraction of histogram of oriented gradient variant (HOGv) feature and 2) a single classifier trained by extreme learning machine (ELM) algorithm. The presented HOGv feature keeps a good balance between redundancy and local details such that it can represent distinctive shapes better. The classifier is a single-hidden-layer feedforward network. Based on ELM algorithm, the connection between input and hidden layers realizes the random feature mapping while only the weights between hidden and output layers are trained. As a result, layer-by-layer tuning is not required. Meanwhile, the norm of output weights is included in the cost function. Therefore, the ELM-based classifier can achieve an optimal and generalized solution for multiclass TSR. Furthermore, it can balance the recognition accuracy and computational cost. Three datasets, including the German TSR benchmark dataset, the Belgium traffic sign classification dataset and the revised mapping and assessing the state of traffic infrastructure (revised MASTIF) dataset, are used to evaluate this proposed method. Experimental results have shown that this proposed method obtains not only high recognition accuracy but also extremely high computational efficiency in both training and recognition processes in these three datasets.
Citations
More filters
Journal Article•DOI•
A review on neural networks with random weights

[...]

Weipeng Cao1, Xizhao Wang1, Zhong Ming1, Jinzhu Gao2•
Shenzhen University1, University of the Pacific (United States)2
31 Jan 2018-Neurocomputing
TL;DR: This paper objectively reviews the advantages and disadvantages of N NRW model, tries to reveal the essence of NNRW, and provides some useful guidelines for users to choose a mechanism to train a feed-forward neural network.

362 citations

Journal Article•DOI•
Deep Learning for Large-Scale Traffic-Sign Detection and Recognition

[...]

Domen Tabernik1, Danijel Skočaj1•
University of Ljubljana1
01 Apr 2020-IEEE Transactions on Intelligent Transportation Systems
TL;DR: A convolutional neural network approach, the mask R-CNN, is adopted to address the full pipeline of detection and recognition with automatic end-to-end learning, which is sufficient for deployment in practical applications of the traffic-sign inventory management.
Abstract: Automatic detection and recognition of traffic signs plays a crucial role in management of the traffic-sign inventory. It provides an accurate and timely way to manage traffic-sign inventory with a minimal human effort. In the computer vision community, the recognition and detection of traffic signs are a well-researched problem. A vast majority of existing approaches perform well on traffic signs needed for advanced driver-assistance and autonomous systems. However, this represents a relatively small number of all traffic signs (around 50 categories out of several hundred) and performance on the remaining set of traffic signs, which are required to eliminate the manual labor in traffic-sign inventory management, remains an open question. In this paper, we address the issue of detecting and recognizing a large number of traffic-sign categories suitable for automating traffic-sign inventory management. We adopt a convolutional neural network (CNN) approach, the mask R-CNN, to address the full pipeline of detection and recognition with automatic end-to-end learning. We propose several improvements that are evaluated on the detection of traffic signs and result in an improved overall performance. This approach is applied to detection of 200 traffic-sign categories represented in our novel dataset. The results are reported on highly challenging traffic-sign categories that have not yet been considered in previous works. We provide comprehensive analysis of the deep learning method for the detection of traffic signs with a large intra-category appearance variation and show below 3% error rates with the proposed approach, which is sufficient for deployment in practical applications of the traffic-sign inventory management.

173 citations

Cites background or methods from "An Efficient Method for Traffic Sig..."

  • ...A method that uses CNN to learn features and then applies ELM as a classifier has been applied in [35], while [36] employed a deep network consisting of spatial transformer layers and a modified version of inception module....

    [...]

  • ...A wide range of machine learning methods have also been employed, ranging from support vector machine (SVM) [24], [16], [27], logistic regression [28], and random forests [16], [27], to artificial neural networks in the form of an extreme learning machine (ELM) [19]....

    [...]

  • ...Traditionally hand-crafted features have been used, like histogram of oriented gradients (HOG) [12], [24], [26], [16], [5], [19], [10], scale invariant feature transform (SIFT) [5], local binary patterns (LBP) [16] or integral channel features [26]....

    [...]

  • ...• The Mapping and Assessing the State of Traffic Infrastructure (MASTIF) [18]: 9 original categories, extended to 31 categories [19], acquired for road maintenance assessment service in Croatia....

    [...]

Journal Article•DOI•
Application of EOS-ELM With Binary Jaya-Based Feature Selection to Real-Time Transient Stability Assessment Using PMU Data

[...]

Yang Li1, Zhen Yang1•
Electric Power University1
24 Oct 2017-IEEE Access
TL;DR: A novel PRTSA approach based on an ensemble of OS-extreme learning machine (EOS-ELM) with binary Jaya (BinJaya)-based feature selection is proposed with the use of phasor measurement units (PMUs) data and has superior computation speed and prediction accuracy than other state-of-the-art sequential learning algorithms.
Abstract: Recent studies show that pattern-recognition-based transient stability assessment (PRTSA) is a promising approach for predicting the transient stability status of power systems. However, many of the current well-known PRTSA methods suffer from excessive training time and complex tuning of parameters, resulting in inefficiency for real-time implementation and lacking the online model updating ability. In this paper, a novel PRTSA approach based on an ensemble of OS-extreme learning machine (EOS-ELM) with binary Jaya (BinJaya)-based feature selection is proposed with the use of phasor measurement units (PMUs) data. After briefly describing the principles of OS-ELM, an EOS-ELM-based PRTSA model is built to predict the post-fault transient stability status of power systems in real time by integrating OS-ELM and an online boosting algorithm, respectively, as a weak classifier and an ensemble learning algorithm. Furthermore, a BinJaya-based feature selection approach is put forward for selecting an optimal feature subset from the entire feature space constituted by a group of system-level classification features extracted from PMU data. The application results on the IEEE 39-bus system and a real provincial system show that the proposal has superior computation speed and prediction accuracy than other state-of-the-art sequential learning algorithms. In addition, without sacrificing the classification performance, the dimension of the input space has been reduced to about one-third of its initial value.

126 citations

Cites background from "An Efficient Method for Traffic Sig..."

  • ...ELM proposed by Huang is a new machine learning approach for single hidden layer feed forward networks [29], and it has been successfully applied in many engineering applications [30-33]....

    [...]

Journal Article•DOI•
Mixture correntropy for robust learning

[...]

Badong Chen, Xin Wang, Na Lu, Shiyuan Wang1, Jiuwen Cao2, Jing Qin3 •
Southwest University1, Hangzhou Dianzi University2, Hong Kong Polytechnic University3
01 Jul 2018-Pattern Recognition
TL;DR: Experimental results show that the learning algorithms under MMCC can perform very well and achieve better performance than the conventional MCC based algorithms as well as several other state-of-the-art algorithms.

124 citations

Cites background from "An Efficient Method for Traffic Sig..."

  • ...Algorithm 1 ELM-MMCC Input: samples {xi, ti}Ni=1 Output: weight vector β Parameters setting: number of hidden nodes Ñ , regularization parameter λ′, maximum iteration number K, kernel width σ1, σ2, mixture coefficient α and termination tolerance ε Initialization: Set β0=0 and randomly initialize the parameters aj and bj ( j = 1, ..., Ñ) 1: for k = 1, 2, ..., K do 2: Compute the error based on βk−1: ei = ti − hiβk−1, i = 1, 2, · · · , N 3: Compute the diagonal matrix Λ: Λii = α σ21 Gσ1 (ei) + 1−α σ22 Gσ2 (ei), i = 1, 2, · · · , N 4: Update the weight vectorβ: βk = [H TΛH + λ′I]−1HTΛ T 5: Until ∣∣JMMCC(βk)− JMMCC(βk−1)∣∣ < ε 6: end for...

    [...]

  • ...The ELM is a kind of singlehidden-layer feedforward neural network (SLFN) with universal approximation capability, whose hidden node parameters are randomly assigned [25, 26, 27, 28, 29, 30, 31, 32, 33]....

    [...]

  • ...Several new learning algorithms are developed under the maximum mixture correntropy criterion (MMCC), including the MMCC based ELM (ELM-MMCC), kernel maximum mixture correntropy (KMMC) and kernel recursive maximum mixture correntropy (KRMMC).35 The remainder of the paper is organized as follows....

    [...]

  • ...Particularly, several robust learning algorithms under the maximum mixture correntropy criterion (MMCC) were developed, including the MMCC based extreme learning machine (ELM-MMCC), kernel maximum mixture correntropy (KMMC) and kernel recursive maximum180 mixture correntropy (KRMMC)....

    [...]

  • ...Clearly, the ELM-MMCC consistently gives superior performance in all cases with only a slight increase125 in execution time....

    [...]

Journal Article•DOI•
Small traffic sign detection from large image

[...]

Zhigang Liu1, Dongyu Li2, Shuzhi Sam Ge3, Shuzhi Sam Ge4, Feng Tian1 •
Northeast Petroleum University1, Harbin Institute of Technology2, Qingdao University3, National University of Singapore4
01 Jan 2020-Applied Intelligence
TL;DR: A deconvolution region-based convolutional neural network (DR-CNN) to cope with small traffic sign detection and a two-stage adaptive classification loss function for region proposal networks (RPN) and fully connected neural networks within DR-CNN is proposed.
Abstract: Automatic traffic sign detection has great potential for intelligent vehicles. The ability to detect small traffic signs in large traffic scenes enhances the safety of intelligent devices. However, small object detection is a challenging problem in computer vision; the main problem involved in accurate traffic sign detection is the small size of the signs. In this paper, we present a deconvolution region-based convolutional neural network (DR-CNN) to cope with this problem. This method first adds a deconvolution layer and a normalization layer to the output of the convolution layer. It concatenates the features of the different layers into a fused feature map to provide sufficient information for small traffic sign detection. To improve training effectiveness and distinguish hard negative samples from easy positive ones, we propose a two-stage adaptive classification loss function for region proposal networks (RPN) and fully connected neural networks within DR-CNN. Finally, we evaluate our proposed method on the new and challenging Tsinghua-Tencent 100K dataset. We further conduct ablation experiments and analyse the effectiveness of the fused feature map and the two-stage classification loss function. The final experimental results demonstrate the superiority of the proposed method for detecting small traffic signs.

87 citations

References
More filters
Journal Article•DOI•
Distinctive Image Features from Scale-Invariant Keypoints

[...]

David G. Lowe1•
University of British Columbia1
01 Nov 2004-International Journal of Computer Vision
TL;DR: This paper presents a method for extracting distinctive invariant features from images that can be used to perform reliable matching between different views of an object or scene and can robustly identify objects among clutter and occlusion while achieving near real-time performance.
Abstract: This paper presents a method for extracting distinctive invariant features from images that can be used to perform reliable matching between different views of an object or scene. The features are invariant to image scale and rotation, and are shown to provide robust matching across a substantial range of affine distortion, change in 3D viewpoint, addition of noise, and change in illumination. The features are highly distinctive, in the sense that a single feature can be correctly matched with high probability against a large database of features from many images. This paper also describes an approach to using these features for object recognition. The recognition proceeds by matching individual features to a database of features from known objects using a fast nearest-neighbor algorithm, followed by a Hough transform to identify clusters belonging to a single object, and finally performing verification through least-squares solution for consistent pose parameters. This approach to recognition can robustly identify objects among clutter and occlusion while achieving near real-time performance.

46,906 citations

Journal Article•DOI•
LIBSVM: A library for support vector machines

[...]

Chih-Chung Chang1, Chih-Jen Lin1•
National Taiwan University1
06 May 2011-ACM Transactions on Intelligent Systems and Technology
TL;DR: Issues such as solving SVM optimization problems theoretical convergence multiclass classification probability estimates and parameter selection are discussed in detail.
Abstract: LIBSVM is a library for Support Vector Machines (SVMs). We have been actively developing this package since the year 2000. The goal is to help users to easily apply SVM to their applications. LIBSVM has gained wide popularity in machine learning and many other areas. In this article, we present all implementation details of LIBSVM. Issues such as solving SVM optimization problems theoretical convergence multiclass classification probability estimates and parameter selection are discussed in detail.

40,826 citations

"An Efficient Method for Traffic Sig..." refers methods in this paper

  • ...In our experiments, SVM and kernel SVM algorithms are implemented using LIBSVM package5 [42]....

    [...]

Proceedings Article•DOI•
Histograms of oriented gradients for human detection

[...]

Navneet Dalal1, Bill Triggs1•
French Institute for Research in Computer Science and Automation1
20 Jun 2005
TL;DR: It is shown experimentally that grids of histograms of oriented gradient (HOG) descriptors significantly outperform existing feature sets for human detection, and the influence of each stage of the computation on performance is studied.
Abstract: We study the question of feature sets for robust visual object recognition; adopting linear SVM based human detection as a test case. After reviewing existing edge and gradient based descriptors, we show experimentally that grids of histograms of oriented gradient (HOG) descriptors significantly outperform existing feature sets for human detection. We study the influence of each stage of the computation on performance, concluding that fine-scale gradients, fine orientation binning, relatively coarse spatial binning, and high-quality local contrast normalization in overlapping descriptor blocks are all important for good results. The new approach gives near-perfect separation on the original MIT pedestrian database, so we introduce a more challenging dataset containing over 1800 annotated human images with a large range of pose variations and backgrounds.

31,952 citations

"An Efficient Method for Traffic Sig..." refers background or methods in this paper

  • ..., histogram of oriented gradients (HOG) [1], are widely used for TSR [2]–[6]....

    [...]

  • ...The HOG variant (HOGv) proposed by this paper has two improvements compared with the original HOG descriptor [1]....

    [...]

  • ..., HOG [1]–[6], scale-invariant feature transform (SIFT) [21]–[23], and Gabor features [24], [25]....

    [...]

Distinctive Image Features from Scale-Invariant Keypoints

[...]

Matthijs Dorst
01 Jan 2011
TL;DR: The Scale-Invariant Feature Transform (or SIFT) algorithm is a highly robust method to extract and consequently match distinctive invariant features from images that can then be used to reliably match objects in diering images.
Abstract: The Scale-Invariant Feature Transform (or SIFT) algorithm is a highly robust method to extract and consequently match distinctive invariant features from images. These features can then be used to reliably match objects in diering images. The algorithm was rst proposed by Lowe [12] and further developed to increase performance resulting in the classic paper [13] that served as foundation for SIFT which has played an important role in robotic and machine vision in the past decade.

14,708 citations

"An Efficient Method for Traffic Sig..." refers background in this paper

  • ...Due to the use of oriented gradients as feature primitives, HOG and SIFT descriptors are both robust to illumination changes....

    [...]

  • ...Furthermore, SIFT feature is robust to rotation and scaling since gradients used for accumulation are relative to the principle orientation and keypoints are extracted based on normalized derivatives (e.g., difference of Gaussian)....

    [...]

  • ...However, keypoint-based SIFT feature is sparse, and therefore, it has to face the problem of inconsistence of dimensionality due to the difference of keypoint numbers between images....

    [...]

  • ...Since traffic signs are salient in terms of shape, features based on statistics in terms of gradient or orientation energy are widely used to represent traffic signs, e.g., HOG [1]–[6], scale-invariant feature transform (SIFT) [21]–[23], and Gabor features [24], [25]....

    [...]

  • ..., HOG [1]–[6], scale-invariant feature transform (SIFT) [21]–[23], and Gabor features [24], [25]....

    [...]

Journal Article•DOI•
Multiresolution gray-scale and rotation invariant texture classification with local binary patterns

[...]

Timo Ojala1, Matti Pietikäinen1, Topi Mäenpää1•
University of Oulu1
01 Jul 2002-IEEE Transactions on Pattern Analysis and Machine Intelligence
TL;DR: A generalized gray-scale and rotation invariant operator presentation that allows for detecting the "uniform" patterns for any quantization of the angular space and for any spatial resolution and presents a method for combining multiple operators for multiresolution analysis.
Abstract: Presents a theoretically very simple, yet efficient, multiresolution approach to gray-scale and rotation invariant texture classification based on local binary patterns and nonparametric discrimination of sample and prototype distributions. The method is based on recognizing that certain local binary patterns, termed "uniform," are fundamental properties of local image texture and their occurrence histogram is proven to be a very powerful texture feature. We derive a generalized gray-scale and rotation invariant operator presentation that allows for detecting the "uniform" patterns for any quantization of the angular space and for any spatial resolution and presents a method for combining multiple operators for multiresolution analysis. The proposed approach is very robust in terms of gray-scale variations since the operator is, by definition, invariant against any monotonic transformation of the gray scale. Another advantage is computational simplicity as the operator can be realized with a few operations in a small neighborhood and a lookup table. Experimental results demonstrate that good discrimination can be achieved with the occurrence statistics of simple rotation invariant local binary patterns.

14,245 citations

Related Papers (5)
Extreme learning machine: Theory and applications

[...]

01 Dec 2006-Neurocomputing
Guang-Bin Huang, Qin-Yu Zhu, Chee Kheong Siew
Extreme Learning Machine for Regression and Multiclass Classification

[...]

01 Apr 2012
Guang-Bin Huang, Hongming Zhou, Xiaojian Ding, Rui Zhang 
Universal approximation using incremental constructive feedforward networks with random hidden nodes

[...]

01 Jul 2006-IEEE Transactions on Neural Networks
Guang-Bin Huang, Lei Chen, Chee-Kheong Siew
Extreme Learning Machine for Multilayer Perceptron

[...]

01 Apr 2016-IEEE Transactions on Neural Networks
Jiexiong Tang, Chenwei Deng, Guang-Bin Huang
Extreme learning machine: a new learning scheme of feedforward neural networks

[...]

25 Jul 2004
Guang-Bin Huang, Qin-Yu Zhu, Chee-Kheong Siew