Advanced scale-space, invariant, low detailed feature recognition from images - car brand recognition

Badura, Stefan; Foltan, Stanislav

doi:10.1109/imcsit.2010.5679924

Cited by 13 publications

(10 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, other features such as traffic lights (front or back) or grille are usually ''hard'' to describe. There are a few scientific papers covering this topic: [5,12,37]. They use the proven histogram of oriented gradients (HOG) and support vector machines (SVM) approach introduced for pedestrian detection in the work [15].…”

Section: Vehicle Type Recognitionmentioning

confidence: 99%

“…The resource usage is summarized in Table 5. 5 The DSP module usage is quite high, due to inside_ROI flag computation. However, in case of insufficient DSP resources, fabric-based multiplier could also be considered.…”

Section: Hardware Implementationmentioning

confidence: 99%

“…To reduce the impact of external lighting conditions and camera jitter, all operations were performed on two 5 All the presented resource usages, if not stated explicitly, are estimations obtained by the ISE software after synthesis. Fig.…”

Section: Vehicle Detection and Countingmentioning

confidence: 99%

See 2 more Smart Citations

Real-time hardware–software embedded vision system for ITS smart camera implemented in Zynq SoC

Kryjak

Komorkiewicz

Gorgon

2016

J Real-Time Image Proc

View full text Add to dashboard Cite

The article demonstrates the usefulness of heterogeneous System on Chip (SoC) devices in smart cameras used in intelligent transportation systems (ITS). In a compact, energy efficient system the following exemplary algorithms were implemented: vehicle queue length estimation, vehicle detection, vehicle counting and speed estimation (using multiple virtual detection lines), as well as vehicle type (local binary features and SVM classifier) and colour (k-means classifier and YCbCr colourspace analysis) recognition. The solution exploits the hardwaresoftware architecture, i.e. the combination of reconfigurable resources and the efficient ARM processor. Most of the modules were implemented in hardware, using Verilog HDL, taking full advantage of the possible parallelization and pipeline, which allowed to obtain real-time image processing. The ARM processor is responsible for executing some parts of the algorithm, i.e. high-level image processing and analysis, as well as for communication with the external systems (e.g. traffic lights controllers). The demonstrated results indicate that modern SoC systems are a very interesting platform for advanced ITS systems and other advanced embedded image processing, analysis and recognition applications.Keywords Intelligent Transportation Systems Á Hardware-software image processing (Zynq SoC) Á Vehicle queue length estimation Á Vehicle detection Á Vehicle type and colour recognition

show abstract

Section: Vehicle Type Recognitionmentioning

confidence: 99%

Section: Hardware Implementationmentioning

confidence: 99%

See 1 more Smart Citation

Real-time hardware–software embedded vision system for ITS smart camera implemented in Zynq SoC

Kryjak

Komorkiewicz

Gorgon

2016

J Real-Time Image Proc

View full text Add to dashboard Cite

show abstract

“…Cai et al [3] used knowledge about the body of an automobile to detect an automobile. Badura and Foltan [4] used SIFT (Scale Invariant Feature Transform) [2] and SURF (Speed up Robust Features) [5] to find interest points and generate an invariant descriptor. Liu Jiamin et al [6] applied Hu invariant moments to the vehicle logo recognition.…”

Section: Introductionmentioning

confidence: 99%

Vehicle Logo Recognition Based on Optimized Dictionary and Robust Collaborative Representation

Li¹,

Huang²,

Cao³

2016

dtetr

View full text Add to dashboard Cite

Abstract. Recently, with the increase of the car ownership, transportation network becomes more and more complex, the traffic management system is facing a serious challenge. Therefore, vehicle recognition plays a decisive role. The vehicle logo contains information of a vehicle, so vehicle logo can be used to identify vehicle models. Considering the problem of the high calculation cost of the vehicle recognition based on sparse representation, a vehicle logo recognition method based on optimized dictionary and robust collaborative representation is proposed. This method not only has a high recognition rate but also reduces the computation time. In our method, HOG feature is selected to represent the vehicle logo image. HOG features are extracted from training samples and testing samples respectively, and then dictionary learning method based on Fisher discrimination criterion is adopted to optimize the train samples. Finally, robust collaborative representation is used to recognize the vehicle models. The experimental results show that our method achieves a high recognition rate 98.2% and reduces the computation time.

show abstract

“…Badura and Foltan [4] used SIFT (Scale Invariant Feature Transorm) [5] and SURF (Speed Up Robust Features) [6], to find interest points and generate an invariant descriptor. The shape of the descriptor was then used to recognize the automobile make from logo image which taken in real environment.…”

Section: Introductionmentioning

confidence: 99%

Vehicle logo recognition based on interior structure using SIFT descriptor and neural network

Lipikorn

Cooharojananone

Kijsupapaisan

et al. 2014

2014 International Conference on Information Science, Electronics and Electrical Engineering

View full text Add to dashboard Cite

This paper presents an approach for automatic recognition of vehicle make from its logo in a front-view image using SIFT descriptor of interior structure and back-propagation neural network. The proposed method focuses on recognition of automobile make by integrating Top-Hat transformation with shape descriptor to locate the logo of an automobile from an image then uses back-propagation neural network to recognize an automobile make from the SIFT (Scale-Invariant Feature Transform) descriptor of inner structure of the logo. The training set contains eighteen images of six different logos, whereas the test set contains 220 images of automobile. The recognition results from the proposed method were compared with the results from other existing methods and the results reveal that it can recognize automobile makes regardless of illumination condition or position and the accuracy rate is over 50%.

show abstract

Advanced scale-space, invariant, low detailed feature recognition from images - car brand recognition

Cited by 13 publications

References 4 publications

Real-time hardware–software embedded vision system for ITS smart camera implemented in Zynq SoC

Real-time hardware–software embedded vision system for ITS smart camera implemented in Zynq SoC

Vehicle Logo Recognition Based on Optimized Dictionary and Robust Collaborative Representation

Vehicle logo recognition based on interior structure using SIFT descriptor and neural network

Contact Info

Product

Resources

About