scispace - formally typeset
Search or ask a question
Proceedings ArticleDOI

A novel method for the extraction of primary visual features from an image through intelligent feature descriptors

TL;DR: The traffic sign images were acquired from the image database and were subjected to some pre-processing techniques such as conversion of the original RGB images into HSV Color Space, Adjustment of the Contrast of the Color images as well as applying the Histogram of Oriented Gradients (HOG) algorithm in which the process of extraction and plotting of the HOG features from a given image is performed that is most popular amongst the feature extraction algorithms.
Abstract: As far as the safety of a driver is concerned, more focus should be put on correct interpretation and information which is conveyed by a traffic sign, while driving a vehicle along the road. A sign board can be thought of as an emblem which disseminates important and meaningful information regarding the potential hazards prevailing among road users comprising roadways cladded with snowfall, construction worksites or repairing of roads taking place and telling the people to follow an alternative route. It alerts the person who is passing through the road about the maximum possible extremity that his vehicle is trying to achieve indicating slowing down the speed of vehicle since chances of having collision cannot be ruled out. With constant increasing of the training database size, not only there cognition accuracy, but also the computation complexity should be considered in designing a feasible recognition approach. The traffic sign images were acquired from the image database and were subjected to some pre-processing techniques such as conversion of the original RGB images into HSV Color Space, Adjustment of the Contrast of the Color images as well as applying the Histogram of Oriented Gradients (HOG) algorithm in which the process of extraction and plotting of the HOG features from a given image is performed that is most popular amongst the feature extraction algorithms. In the future, we will concentrate on detecting, recognizing as well as classifying a particular sign board.
Citations
More filters
Proceedings ArticleDOI
01 Aug 2018
TL;DR: The traffic sign images were acquired from the image database and were subjected to some pre-processing techniques such as removing the noise present in a particular image with the help of Arithmetic Mean Filter as well as Geometric Mean Filter.
Abstract: As far as the safety of a driver is concerned, more focus should be put on correct interpretation and information which is conveyed by a traffic sign, while driving a vehicle along the road. A sign board can be thought of as an emblem which disseminates important and meaningful information regarding the potential hazards prevailing among road users comprising roadways cladded with snowfall, construction worksites or repairing of roads taking place and telling the people to follow an alternative route. It alerts the person who is passing through the road about the maximum possible extremity that his vehicle is trying to achieve indicating; slowing down the speed of vehicle since chances of having collision cannot be ruled out. With constant increasing of the training database size, not only the recognition accuracy, but also the computation complexity should be considered in designing a feasible recognition approach. The traffic sign images were acquired from the image database and were subjected to some pre-processing techniques such as removing the noise present in a particular image with the help of Arithmetic Mean Filter as well as Geometric Mean Filter. In the future, we will concentrate on detecting, recognizing as well as classifying a particular sign board.

Cites background or methods from "A novel method for the extraction o..."

  • ...The circular sign board images are detected by making use of the colorbased and/or the shape-based methods, for example, if we consider the research work which is carried out by [21][22][23][24][25]....

    [...]

  • ...There is an urgent need for incorporating an automatic system for the purpose of recognizing the sign boards and is considered to be a main factor for building an autonomous navigation system [19][22][23][24][25]....

    [...]

  • ...An SVM segmentation approach for traffic sign recognition was given in [17][19][22][23][24], while in [18][19][22][23][25], an effective strategy helping in the process of recognizing slanted speed limit signs by extracting the rotation invariant features with the help of Fourier based wavelet descriptor was introduced....

    [...]

  • ...If we take the example of digit-based methods, the process of extracting multiple characters for recognizing numerical portrait of velocity which is exhibited by a moving vehicle rather than classifying a complete sign board image treating it as a single entity is performed [18][22][24][25]....

    [...]

  • ...rectangular sign board images [15][18][21][22][24][25]....

    [...]

References
More filters
Journal ArticleDOI
TL;DR: A novel approach for the detection and classification of traffic signs that offers high performance and better accuracy than the state-of-the-art strategies and is potentially better in terms of noise, affine deformation, partial occlusions, and reduced illumination.
Abstract: The high variability of sign appearance in uncontrolled environments has made the detection and classification of road signs a challenging problem in computer vision. In this paper, we introduce a novel approach for the detection and classification of traffic signs. Detection is based on a boosted detectors cascade, trained with a novel evolutionary version of Adaboost, which allows the use of large feature spaces. Classification is defined as a multiclass categorization problem. A battery of classifiers is trained to split classes in an Error-Correcting Output Code (ECOC) framework. We propose an ECOC design through a forest of optimal tree structures that are embedded in the ECOC matrix. The novel system offers high performance and better accuracy than the state-of-the-art strategies and is potentially better in terms of noise, affine deformation, partial occlusions, and reduced illumination.

195 citations


"A novel method for the extraction o..." refers background in this paper

  • ...It can be very well understood from this study that a comparatively greater emphasis was laid in enhancing the accuracy of existing methodologies prevalent in detecting and recognizing sign boards in an automatic fashion thereby resulting in the reduction of number of support vectors which are required in due course of the complete process leading to a suddenfall in the necessity of memory and time for testing new samples [14][19][20][22]....

    [...]

Journal ArticleDOI
TL;DR: A novel framework with two deep learning components including fully convolutional network (FCN) guided traffic sign proposals and deep Convolutional neural network (CNN) for object classification to perform fast and accurate traffic sign detection and recognition.

181 citations


"A novel method for the extraction o..." refers background or methods in this paper

  • ...An SVM segmentation approach for traffic sign recognition was given in [10][12][16][19], while in [14][17][21][25], an effective strategy helping in the process of recognizing slanted speed limit signs by extracting the rotation invariant features with the help of Fourier based wavelet descriptor was introduced....

    [...]

  • ...An entirely different approach that is suitable for classifying sign boards belonging to different classes by using the ECOC technique was introduced in [2][4][6][8][12]....

    [...]

  • ...One of the major challenging tasks which is confronted by modern car manufacturing companies is to recognize the sign boards correctly particularly in an uncontrolled environment [9] [10] [12] [15]....

    [...]

  • ...By using the concept of a class similarity measure that is learnt from image pairs, realization was achieved with the help of a new newer version of the existing AdaBoost algorithm which is known as SimBoost algorithm [5][7][8][10][12]....

    [...]

  • ...In [8][12][14][16][17], a comparison was performed taking into account the class specific sets of discriminative local regions in between a discrete color image of the viewed sign with other images....

    [...]

Journal ArticleDOI
01 Sep 2016
TL;DR: A new traffic sign detection and recognition method, which is achieved in three main steps, to use invariant geometric moments to classify shapes instead of machine learning algorithms and the results obtained are satisfactory when compared to the state-of-the-art methods.
Abstract: Graphical abstractDisplay Omitted In this paper we present a new traffic sign detection and recognition (TSDR) method, which is achieved in three main steps. The first step segments the image based on thresholding of HSI color space components. The second step detects traffic signs by processing the blobs extracted by the first step. The last one performs the recognition of the detected traffic signs. The main contributions of the paper are as follows. First, we propose, in the second step, to use invariant geometric moments to classify shapes instead of machine learning algorithms. Second, inspired by the existing features, new ones have been proposed for the recognition. The histogram of oriented gradients (HOG) features has been extended to the HSI color space and combined with the local self-similarity (LSS) features to get the descriptor we use in our algorithm. As a classifier, random forest and support vector machine (SVM) classifiers have been tested together with the new descriptor. The proposed method has been tested on both the German Traffic Sign Detection and Recognition Benchmark and the Swedish Traffic Signs Data sets. The results obtained are satisfactory when compared to the state-of-the-art methods.

137 citations


"A novel method for the extraction o..." refers background or methods in this paper

  • ...A color insensitive Haar wavelets feature combined with the AdaBoost algorithm was introduced in to develop a country-independent recognition module [11][13][15][16][19]....

    [...]

  • ...192 were learnt in offline mode from the standard templates of sign board images [11][13][15][16][19]....

    [...]

  • ...In [11][13][14][16], a shape-based classification was developed using an SVM....

    [...]

Journal ArticleDOI
TL;DR: The objectives of this work are to propose pre-processing methods and improvements in support vector machines to increase the accuracy achieved while the number of support vectors, and thus theNumber of operations needed in the test phase, is reduced.

124 citations

Journal ArticleDOI
TL;DR: This work shows how a combination of solid image analysis and pattern recognition techniques can be used to tackle the problem of traffic sign detection in mobile mapping data, and presents in detail the design of a Traffic Sign Detection pipeline.

105 citations


"A novel method for the extraction o..." refers background or methods in this paper

  • ...An entirely different approach that is suitable for classifying sign boards belonging to different classes by using the ECOC technique was introduced in [2][4][6][8][12]....

    [...]

  • ...By using the concept of a class similarity measure that is learnt from image pairs, realization was achieved with the help of a new newer version of the existing AdaBoost algorithm which is known as SimBoost algorithm [5][7][8][10][12]....

    [...]

  • ...In [8][12][14][16][17], a comparison was performed taking into account the class specific sets of discriminative local regions in between a discrete color image of the viewed sign with other images....

    [...]

  • ...The area of Intelligent Transportation Systems is gaining popularity nowadays and many car manufacturing companies are focusing their eyes on the area of Advanced Driver Assistance Systems since it contains a broad spectrum available for carrying out research and development especially in the domain of Traffic Sign Recognition [4][5] [8]....

    [...]

  • ...The results which were obtained from the various classifier algorithms were combined over time and merging of the signs that are quite similar on both the sides of the roadway was accomplished taking help of fusion module that incorporated combining a Bayesian network and a decision tree [6][8][10][12][14]....

    [...]