Artificial Neural Network-Based Multifocus Image Fusion
By: Manchandana, Meenu.
Contributor(s): Gambhir, Deepak.
Publisher: New Delhi STM Journals 2018Edition: Vol 5 (3), Sep -Dec.Description: 16 - 23p.Subject(s): Computer EngineeringOnline resources: Click Here In: Journal of artificial intelligence research and advances (JoAIRA)Summary: Fusion of multiple images is needed for combining information contained in images obtained using different imaging devices or different camera settings. Due to the physics of optical lenses present in these devices, it is difficult to capture an image that has all the relevant objects well in focus. Thus, artificial neural network (ANN)-based multifocus image fusion method in YCbCr color space is proposed. Input images are first mapped from RGB color space to YCbCr color space. Y-component of both input images is then divided into same-size blocks. Features that reflect the clarity of these blocks are extracted and used by ANN to select the input blocks that are used to produce a single all-in-focus fused image. The fused image is then mapped back into the RGB color space, thereby producing the final fused image. Both subjective and objective results prove that the proposed ANN-based multifocus image fusion method outperforms the traditional wavelet transforms-based image fusion algorithms. Cite this ArticleItem type | Current location | Call number | Status | Date due | Barcode | Item holds |
---|---|---|---|---|---|---|
Articles Abstract Database | School of Engineering & Technology Archieval Section | Not for loan | 2021-2021418 |
Fusion of multiple images is needed for combining information contained in images obtained using different imaging devices or different camera settings. Due to the physics of optical lenses present in these devices, it is difficult to capture an image that has all the relevant objects well in focus. Thus, artificial neural network (ANN)-based multifocus image fusion method in YCbCr color space is proposed. Input images are first mapped from RGB color space to YCbCr color space. Y-component of both input images is then divided into same-size blocks. Features that reflect the clarity of these blocks are extracted and used by ANN to select the input blocks that are used to produce a single all-in-focus fused image. The fused image is then mapped back into the RGB color space, thereby producing the final fused image. Both subjective and objective results prove that the proposed ANN-based multifocus image fusion method outperforms the traditional wavelet transforms-based image fusion algorithms.
Cite this Article
There are no comments for this item.