Semantic Tolerance Relation-Based Image Representation and Classification

Lee, Woon-Gu; Chae, Cheol-Joo; Kim, Sang-Kuk; Kang, Mu-Yeong; Lee, Jae-Kwang

doi:10.1109/ipc.2007.26

Cited by 3 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, only the binary classes (indoor-outdoor, manmade-natural, sunsetnon-sunset) were handled. In [5], a method of semantic tolerance relation-based image presentation and classification was proposed, and as the demonstration, the semantics of each image regarding nature vs. man-made domain was represented by assigning 7 categories to image based on the Bayesian classifier. Obviously, it is not enough to embody the whole semantics of images set only by 7 defined categories.…”

Section: Introductionmentioning

confidence: 99%

Representing Meanings of Images Based on Associative Values with Lexicons

Dai

2009

2009 Fifth International Conference on Intelligent Information Hiding and Multimedia Signal Processing

Self Cite

View full text Add to dashboard Cite

In this paper, an approach of representing meanings of images based on associative values with lexicons is proposed. For this, the semantic tolerance relation model (STRM) that reflects the tolerance degree between defined lexicons is generated, and two factors of semantic relevance (SR) and visual similarity (VS) are considered in generating associative values. Furthermore, the algorithm of calculating associative values using bidirectional associative memories (BAM), which is easy to implement, is depicted. The experiment results show that our proposed method improves the accuracy of retrieving images because of the introduction of SR and VS in representing meanings of images.

show abstract

Section: Introductionmentioning

confidence: 99%

Representing Meanings of Images Based on Associative Values with Lexicons

Dai

2009

2009 Fifth International Conference on Intelligent Information Hiding and Multimedia Signal Processing

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, only the binary classes (indoor-outdoor, manmade-natural, sunset-non-sunset) were handled. In [5], a method of semantic tolerance relation-based image presentation and classification was proposed, and as the demonstration, the semantics of each image regarding nature vs. man-made domain was represented by assigning 7 categories to image based on the Bayesian classifier. Obviously, it is not enough to embody the whole semantics of images set only by 7 defined categories.…”

Section: Introductionmentioning

confidence: 99%

“…Because most image/videos have multiple semantic interpretations and people judge similar images with different criterion, in this paper, on the basis of the semantic tolerance relation model (STRM) in respect of image semantics presented in [5], we propose a meta-data structure to represent the image or key frame of the video by the associative values with the defined dimensions and classes, and describe the method how to generate the associative values automatically for different dimensions. Especially, for the nature vs. manmade dimension, in the case of generating associative values by the bidirectional associative memory, the principle of defining the categories and selecting stored pattern images for the bidirectional associate memory is explained.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Image/ video's automatic annotation considering semantics' tolerance relation

Dai

2008

2008 IEEE International Conference on Systems, Man and Cybernetics

Self Cite

View full text Add to dashboard Cite

In this paper, an approach of image/video's automatic annotation considering semantics' tolerance relation between pre-defined classes is proposed. The image/videos' semantics is represented by a vector of associative values with the pre-defined classes, and eliminating the semantic gap induced in the process of automatically generating associative values is dealt with by using the constructed semantic tolerance relation model (STRM). Moreover, the method how to generating associative values with pre-defined classes regarding different dimensions is described. Image/videos' retrieval results based on the predefined classes shows the effectiveness of the proposed methods.

show abstract