Object recognition based on ORB and self-adaptive kernel clustering algorithm

Fuzzy kernel clustering algorithm is a combination of unsupervised clustering and fuzzy set of the concept of image segmentation techniques, But the algorithm is sensitive to initial value, to a large extent dependent on the initial clustering center of choice, and easy to converge to local minimum values, when used in image segmentation, membership of the calculation only consider the current pixel values in the image, and did not consider the relationship between neighborhood pixels, and so on segmentation contains noise image is not ideal. This paper puts forward an improved fuzzy kernel clustering image segmentation algorithm, the multi-objective problem, change the single objective problem to increase the secondary goals concerning membership functions, Then add the constraint information space; Finally, using spatial neighborhood pixels corrected membership degree of the current pixel. The experimental results show that the algorithm effectively avoids the algorithm converges to local extremism and the stagnation of the iterative process will appear problem, significantly lower iterative times, and has good robustness and adaptability.

Download Full-text

Improved Object Recognition with Decision Trees Using Subspace Clustering

Journal of Advanced Computational Intelligence and Intelligent Informatics ◽

10.20965/jaciii.2016.p0041 ◽

2016 ◽

Vol 20 (1) ◽

pp. 41-48 ◽

Cited By ~ 1

Author(s):

Billy Peralta ◽

◽

Luis Alberto Caro

Keyword(s):

Object Recognition ◽

Decision Trees ◽

Clustering Algorithm ◽

Comprehensive Evaluation ◽

Recognition Performance ◽

Clustering Algorithms ◽

Subspace Clustering ◽

Visual Words ◽

Ensemble Techniques ◽

Standard Object

Generic object recognition algorithms usually require complex classificationmodels because of intrinsic difficulties arising from problems such as changes in pose, lighting conditions, or partial occlusions. Decision trees present an inexpensive alternative for classification tasks and offer the advantage of being simple to understand. On the other hand, a common scheme for object recognition is given by the appearances of visual words, also known as the bag-of-words method. Although multiple co-occurrences of visual words are more informative regarding visual classes, a comprehensive evaluation of such combinations is unfeasible because it would result in a combinatorial explosion. In this paper, we propose to obtain the multiple co-occurrences of visual words using a variant of the CLIQUE subspace-clustering algorithm for improving the object recognition performance of simple decision trees. Experiments on standard object datasets show that our method improves the accuracy of the classification of generic objects in comparison to traditional decision tree techniques that are similar, in terms of accuracy, to ensemble techniques. In future we plan to evaluate other variants of decision trees, and apply other subspace-clustering algorithms.

Download Full-text