Authors: Asha Rajbhoj, Jayavardhana Gubbi, P. Balamuralidhar, Shailesh Deshpande, Vinay Kulkarni
Tags: 2019, conceptual modeling
Creation of an image processing pipeline for solving complex problems is a tedious task. Current industry practices largely rely on the image processing domain experts for this. Given the image processing problem have multiple viable solutions. Thus, the search space of creating suitable solution using available algorithms for a given goal in a given constrained infrastructure is generally large. The exploratory work to choose an optimal image processing solution is an effort-, time- and intellect-intensive endeavor. To address these issues we propose a system for automatic construction of the pipeline that can improve domain expert’s productivity by creating a solution quickly. The proposed system externalizes image processing domain knowledge in the form of object model and a set of rules defined over it. Recommendations are given to choose suitable algorithm/s for carrying out the image processing tasks. On successful creation of the pipeline, the system generates deployable code. It also generates trace data that can help for cognitive knowledge upgrade. We showcase ongoing work on this system and its early results using the simple working example.Read the full paper here: https://link.springer.com/chapter/10.1007/978-3-030-20618-5_20
