In this paper, we report formulation and case study of the conditions for patenting processes of content-based retrieval in digital libraries, especially in image libraries. Inventors and practitioners demand formulation of the conditions for patenting the processes as computer-related programs in combining prior disclosed means, and also in comprising the means for parameter settings to perform certain functions. Content-based retrieval indexes the extracted features of images and classifies the indexes to perform its retrieval function.A process for content-based retrieval often consists of a combination of prior disclosed means. That process also comprises the means for parameter settings that are adjusted to retrieve a specific kind of image at a certain narrow domain. We formulate the conditions of patentability on processes for performing content-based retrieval in combining the prior disclosed means and/or comprising the means for parameter settings from the practical standpoints of technical advancement (nonobviousness) and specification (enablement).
CITATION STYLE
Sasaki, H., & Kiyoki, Y. (2002). Patenting the processes for content-based retrieval in digital libraries. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2555, pp. 471–482). Springer Verlag. https://doi.org/10.1007/3-540-36227-4_57
Mendeley helps you to discover research relevant for your work.