Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{200723,
        author = {Gagan Murthy and Abhinav Dinkar and Hitha R Shetty and Manoj B and Soni M},
        title = {Efficiency and Reliability Improvement in a Raspberry Pi Based Pipeline Inspection Robot},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {1330-1338},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=200723},
        abstract = {Pipeline inspection robots are commonly used to detect cracks and structural defects inside industrial pipelines. In low-cost embedded systems, deep-learning models can create high computational load and unstable detection outputs, reducing both efficiency and reliability. This work evaluates a Raspberry Pi based vision inspection robot and applies two deployment-level optimizations: INT8 quantization of a baseline FP32 YOLO model and multi-frame confidence confirmation. Quantization reduces inference latency and CPU usage, while temporal confirmation reduces false positives caused by single-frame instability. Results from controlled experiments show substantial runtime improvement with minor confidence reduction and improved detection stability during continuous inspection. The combined method provides a practical balance between efficiency and reliability for low-cost embedded pipeline inspection.},
        keywords = {Detection reliability, Edge AI, INT8 quantization, Pipeline inspection, Raspberry Pi, YOLO.},
        month = {May},
        }