Please use this identifier to cite or link to this item:
http://hdl.handle.net/2080/4095
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Panda, Santosh Kumar | - |
dc.contributor.author | Nayak, Devidutta | - |
dc.contributor.author | Sa, Pankaj Kumar | - |
dc.date.accessioned | 2023-11-17T11:19:15Z | - |
dc.date.available | 2023-11-17T11:19:15Z | - |
dc.date.issued | 2023-11 | - |
dc.identifier.citation | 8th International Conference On Computer Vision and Image Processing (CVIP) 2023 At IIT Jammu During 3rd -5th November 2023 | en_US |
dc.identifier.uri | http://hdl.handle.net/2080/4095 | - |
dc.description | Copyright belongs to proceeding publisher | en_US |
dc.description.abstract | Low-light images are captured either during nighttime or in environments with limited illumination. These images typically suffer from reduced visibility and increased noise levels, making it challenging to extract relevant information. Enhancing such images is a decisive task in many fields, like medical imaging and night surveillance, where the clarity and detailing of the image are required for better image processing. This article introduces a novel deep-learning-based approach for enhancing low-light images. Our technique utilizes a multiscale cross-connection network (MCCNet) that employs region-specific analysis to improve feature extraction. Moreover, our method also includes a calibrated network to fine-tune the features to get improved results. The critical section of the model is the region selection of the images, where some overlapping regions help provide contextual information between the extracted features. The cross-connection network helps to share the distinct features extracted and concatenate them to a single feature block, further improving the model’s performance. Substantial simulations show our model’s haughty performance compared with the state-of-the-art benchmark methods. The GitHub code is available at https://github.com/santoshpanda1995/Multiscale-crossconnection-network | en_US |
dc.subject | Low-light image enhancement (LLIE) | en_US |
dc.subject | Multi-scale Regions | en_US |
dc.subject | Cross-connection | en_US |
dc.subject | Calibrated-CNN | en_US |
dc.title | MCCNet: A Multi-Scale Cross Connection Network for Low-Light Image Enhancement | en_US |
dc.type | Article | en_US |
Appears in Collections: | Conference Papers |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
2023_CVIP_SKPanda_MCCNet.pdf | 13.75 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.