Artificial intelligence processing for enhancing an intelligent sensor - II

• 主要作者: Potipireddi Sai Pratyusha
• 其他作者: Zheng Yuanjin
• 格式: Final Year Project
• 語言: English
• 出版: Nanyang Technological University 2023
• 主題: Engineering::Electrical and electronic engineering
• 在線閱讀: https://hdl.handle.net/10356/167049

This final year project focused on the objective of finding an Artificial Intelligence algorithm for reconstructing photoacoustic images. The theory behind the algorithms used, their implementation and results were discussed. First, the significance of this project was explained through the need for a non-invasive blood pressure measurement technique such as Photoacoustic Imaging. The working principle of a Photoacoustic Microscopy System was explained along with the conversion of the signals to images. The enhancement of the image details using convolutional neural networks like U-Net and Generative Adversarial Network was introduced. The network architectures and the advantages were highlighted. The two networks were constructed, and an input dataset of 1000 preprocessed photoacoustic images was used. The models built were evaluated based on the test dataset using metrics such as Peak signal-to-noise ratio and Structural Similarity Index Measure. The output of the models that are the reconstructed images are also compared. The best reconstruction algorithm was proposed based on the results obtained.