A Specular Reflection Removal Technique in Cervigrams
- Resource Type
- Conference
- Authors
- Mukku, Lalasa; Thomas, Jyothi
- Source
- 2023 IEEE International Conference on Contemporary Computing and Communications (InC4) Contemporary Computing and Communications (InC4), 2023 IEEE International Conference on. 1:1-5 Apr, 2023
- Subject
- Communication, Networking and Broadcast Technologies
Computing and Processing
Robotics and Control Systems
Signal Processing and Analysis
Deep learning
Image segmentation
Sensitivity
Lighting
Reflection
Lesions
Indexes
specular reflections
cervix
preprocessing
HSI images
specular inpainting
- Language
Cancer detection through medical image segmentation and classification is possible owing to the advancement in image processing techniques. Segmentation and classification tasks carried out to predict and classify diseases need to be dependable and precise. Specular reflections are the high-intensity and low-saturation areas that reflect the light from the probing devices that capture the picture of the organ surface. These areas sometimes mimic the features that are key identifying factors for cancers like acetowhite lesions. This review article examines the various methods proposed for removing specular reflections from medical images, especially those captured by colposcopes. The fundamentals of specular reflection removal and its associated challenges are discussed. The paper reviews several prominent approaches for removal of specular reflections proposes a novel method to remove the specular reflections. The comprehensive review can be a strong foundation for researchers looking to decide on appropriate techniques to employ in their respective research approaches.