No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Data structure and retrieval method of scenic image database basedon fuzzy set theory
Abstract
In this paper, the data structure and information retrieval are discussed for the management of an image database for storing images of sketches, pictures, paintings, etc. Because their attributes are very much fuzzy, the authors describe their fuzzy attributes in fuzzy sets and their membership functions. Corresponding to the fuzzy description, the grade of satisfaction of the information retrieval is defined by both retrieval condition and its complement. This model is suitable for the storage of fuzzy data
In this paper, a new data augmentation algorithm, named Mask2Defect is proposed. Via prior knowledge based data infusing, this method is able to generate defects with varied features. Large volume of defects with different shapes, severities, scales, rotation angles, spatial locations, and part numbers can be generated in a controllable manner. These generated defects will work as teacher samples to fine-tune the inspection model, and automatically adapt it to a wider range of defects. To be specific, we first encode the prior knowledge into the teacher mask via the Industrial Prior Knowledge Encoder, and render the defect details according to the mask with the Mask-to-Defect Construction Network. Then, the Fake-to-Real Domain Transformation GAN is used to transform the rendered samples from the fake domain into the real defect domain. Experiments reveal that the synthesized image quality of our method outperforms the state-of-the-art generative methods, and the performance of the inspection model in defect classification and localization has also been improved by fine-tuned with the generated samples.
The composition of a painting consists of colors and shapes, and
is explained with impression words and location of objects. These cause
our impression or feeling. Especially from the viewpoint of KANSEI
information retrieval, the data structure should be designed to be
consistent with the attributes of color, shape, and impression words.
Also, in the information retrieval of paintings, bibliographic data are
important factors, as well as color and shape. The authors discuss the
design and structure of keywords and key images as attributes of
paintings for information retrieval. Each attribute of a painting is
described as a fuzzy set and its membership function that represents the
grade of the keywords or key images. We propose an estimation method for
retrieved objects with the membership function. Furthermore, the
information retrieval of paintings is formulated based on fuzzy set
theory as an extension of the traditional crisp model
Fuzzy Information Retrieval for Scenic Inage Database
- Y Isoioto
- K Yoshine
- H Nakatani Ant
- N Ishii