Color image segmentation using competitive learning

We describe here a novel defect classification system that works in real-time on the images of material running on the production line, provided by a video-inspection module. The classifier is constituted of a two-levels hierarchical architecture: a set of adequate features are extracted from the image first; the defect type is then identified from them. Statistical analysis allows greatly reducing the number of features, leaving only the most significant ones. A particular version of multi-class boosting has been developed for labeling: differently from its original version, it accepts only one label for each image, the true one, and does not require the rank for the other classes. Nevertheless, the classifier is able to produce a valid rank of the defect with respect to all the classes, information that can be used to achieve an identification rate of the dangerous defects very close to 100 % on a real data set.

show abstract

“…To assess the degree of uniformity, the image inside B B is clustered into two or three disjointed regions ( [44], cf. Fig.…”

Section: Regular Defectsmentioning

confidence: 99%

Automatic Defect Classification on a Production Line

Borghese

Fomasi

2015

Intell Ind Syst

View full text Add to dashboard Cite

show abstract

“…One of the early applications of the use of ANNs in medical image segmentation was by Ozkan et al, who used the backpropagation learning to segment MR images [6]. Uchiyama and Arbib used competitive learning to cluster colors in images [7]. Littmann and Ritter developed an ANN, named local linear maps, for adaptive color segmentation and compared it with statistical methods [8].…”

Section: Introductionmentioning

confidence: 99%

Image segmentation by relaxation using constraint satisfaction neural network

Kurugöllü

Sankur

Harmanci

2002

Image and Vision Computing

View full text Add to dashboard Cite

“…Similarly, Lin et al [13] also applied a competitive Hop-"eld neural network to demonstrate the promising results in medical image segmentation. Furthermore, Uchiyama and Arbib [14] used competitive learning as an e$cient method in color image segmentation application. The winner-take-all rule employed by the competitive learning mechanism ensures that only one job is executed on a dedicated processor at a certain time, forcing the 1-out-of-N constraint to be held.…”

Section: Introductionmentioning

confidence: 99%

Competitive neural network to solve scheduling problems

Chen

Huang

2001

Neurocomputing

View full text Add to dashboard Cite

Most scheduling problems have been demonstrated to be NP-complete problems. The Hop"eld neural network is commonly applied to obtain an optimal solution in various di!erent scheduling applications, such as the traveling salesman problem (TSP), a typical discrete combinatorial problem. Hop"eld neural networks, although providing rapid convergence to the solution, require extensive e!ort to determine coe$cients. A competitive learning rule provides a highly e!ective means of attaining a sound solution and can reduce the e!ort of obtaining coe$cients. Restated, the competitive mechanism reduces the network complexity. This important feature is applied to the Hop"eld neural network to derive a new technique, i.e. the competitive Hop"eld neural network technique. This investigation employs the competitive Hop"eld neural network to resolve a multiprocessor problem with no process migration, time constraints (execution time and deadline), and limited resources. Simulation results demonstrate that the competitive Hop"eld neural network imposed on the proposed energy function ensures an appropriate approach to solving this class of scheduling problems.

show abstract

Color image segmentation using competitive learning

Cited by 193 publications

References 19 publications

Automatic Defect Classification on a Production Line

Automatic Defect Classification on a Production Line

Image segmentation by relaxation using constraint satisfaction neural network

Competitive neural network to solve scheduling problems

Contact Info

Product

Resources

About