Hybrid Fuzzy Colour Processing and Learning

Playne, Daniel P.; Mehta, Vrushank D.; Reyes, Napoleon H.; Barczak, Andre L. C.

doi:10.1007/978-3-540-69162-4_40

Cited by 3 publications

(11 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In 2006 [17], FCCF was combined with Adaboost training for automatic colour classification that does not rely on the pieslice colour classifier. Most recently, in 2008 [6], FCCF was fused with a colour learning algorithm that utilises successive frames containing the object being tracked for automatic calibration of the bounding angles used by the pie-slice classifier. In addition, a colour classification scoring system was proposed and was tested to successfully extract the best colour contrast rule combination.…”

Section: Related Workmentioning

confidence: 99%

“…However, the colour clusterings are unstable even when the object being tracked down is under a fixed illumination because quantum electrical effects in the camera sensor chip easily distorts the colours captured and cannot be prevented. Even worse, the colour clusters drift in the colour space due to spatially varying illuminations [2,[5][6][7][8][9][10]. Figure 1 illustrates a snapshot of solid colour patches under spatially varying illuminations.…”

Section: Problem Domainmentioning

confidence: 99%

“…The main contributions of this paper is then discussed in Sect. 6. It provides details on the development of the novel Heuristic-Assisted Genetic Algorithm (HAGA)and its integration with FCCF: the architecture, chromosome design, fitness function and issues on discretisation of the calibration search space.…”

Section: Introductionmentioning

confidence: 99%

“…Finding the pie-slice classification region originally depends on manual calibration and is a painstaking process. In [6], a motion-based colour learning technique was proposed to extract the pie-slice classification region, but the extraction of the optimal colour contrast operation rules was performed only through a brute-force search method. What is desirable is an automatic colour classification system that searches for the pie-slice classification region and colour contrast rules more effectively.…”

Section: Introductionmentioning

confidence: 99%

“…The colour classification process we refer to here pertains to correctly distinguishing colours belonging to the same object as it moves across the scene, with the presence of other similar colours. Although there are many researches about colour classification, only few researches successfully classify colours under varying illuminations [2,[5][6][7][8][9][10].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Finding near optimum colour classifiers: genetic algorithm-assisted fuzzy colour contrast fusion using variable colour depth

Shin

Reyes

2009

Memetic Comp.

Self Cite

View full text Add to dashboard Cite

This paper presents a complete Fuzzy-Geneticbased self-calibrating illumination intensity-invariant colour classification system. Previously, we have developed a novel fuzzy colour processing technique called Fuzzy Colour Contrast Fusion (FCCF) that selectively and adaptively corrects colours depicting target colour objects. FCCF has been proven to compensate for the effects of spatially varying illumination intensities in the scene, in various colour spaces. However, FCCF requires a huge set of parameters that is extremely tedious to calibrate by hand. To address these problems, we present a system that combines FCCF with a Heuristic-Assisted Genetic Algorithm (HAGA). FCCF-HAGA fully automates the fine-tuning of all colour descriptors, with significantly improved colour classification accuracy. Furthermore, we have reduced FCCF's storage space requirements by processing colour channels selectively at varying colour depths. This is accomplished by combining a Variable Colour Depth (VCD) algorithm with FCCF that searches for the most effective colour depth for each colour channel. Our results show that for all cases, the FCCF-HAGA-VCD combination improves pie-slice colour classification. For six different target colours, under varying illuminations, the hybrid algorithm was able to yield 17.63% higher overall colour classification accuracy as compared to the pure fuzzy approach. Furthermore, it was able to reduce LUT storage space requirements by 78.06%, as compared to the full-colour depth LUT.

show abstract

Section: Related Workmentioning

confidence: 99%