Recognizing traffic signs using flexible Discrete Cosine Transform (DCT) grid

Noon, Serosh Karim; Javed, Kashif; Mannan, Abdul; Babri, Haroon A.

doi:10.24200/sci.2017.4121

Cited by 1 publication

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ability of discrete cosine transform (DCT) for feature extraction in images has already been explored [41,42,2,43]. Ayyalasomayajula et al [42] proposed DCT phase (i.e., sign, either positive or negative) based descriptor to recognize partially occluded real world images.…”

Section: Multiscale Spectral Feature Extractionmentioning

confidence: 99%

“…Mainly, there are two types of traffic signs i.e. (1) American with white background and black foreground and (2) European having either red rim or filled with blue color [2,3,4,5]. In this work, we focus on European traffic signs only.…”

Section: Introductionmentioning

confidence: 99%

“…and higher order spectra [16]. Feature extraction can be applied on an image as a whole or with the help of a rectangular or polar grid [2]. Reducing the size of feature vector by getting rid of irrelevant and redundant features, dimensionality reduction [17] has been proved to be an effective method.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Cognition based Recognition of Partially Occluded Traffic Signs

et al. 2022

View full text Add to dashboard Cite

Computer vision based traffic sign detection and recognition is an active field of research but the task becomes challenging when the sign of interest is partially occluded by nearby objects like a tree, pole or vehicle. Another difficulty posed especially in the developing countries is the lost colors problem that arises due to aging and poor maintenance. This work presents an automatic technique that focuses on visible parts only and suppresses occluded portions. Features are collected using a convolutional neural network inspired invariant feature extraction technique augmented with feature interaction based dimensionality reduction. Further, with the use of dynamic parameter estimation, an adaptive system for continuous learning is also proposed. Since the effect of partial occlusion has not been thoroughly studied, there is no benchmark database available for this purpose. We have prepared two datasets by combining originally and synthetically occluded images taken from field surveys and from famous GTSRB database. Experiments revealed that our technique outperforms state of the art recognition methods previously used for visible and occluded signs by obtaining 0.81 precision and 0.79 recall values on the average. The proposed method also shows a remarkably low error rate as the amount of occlusion is increased.

show abstract