Automatic glare removal in reflectance imagery of the uterine cervix

Abstract

Colposcopy is a diagnostic method used to detect cancer precursors and cancer of the uterine cervix. Computer-AidedDiagnosis (CAD) for colposcopy is a new field in medical image processing. Colposcopists analyze glare (glint or specular reflection) patterns on the cervix to assess the surface contour (3D topology) of lesions, an important feature used to evaluate lesion severity. However, glare in the imagery presents major problems for automated image analysis systems. Glare eliminates all information in affected pixels and can introduce artifacts in feature extraction algorithms, such as acetowhite region detection. Although cross-polarization filters can be used to eliminate glare, the reality is that we have to deal with glare when we want to use existing cervical image databases or use an instrument that does not provide cross-polarized imagery. Here, we present the details and preliminary results of a glare removal algorithm for RGB color images of the cervix that can be used as a pre-processing step in CAD systems. The algorithm can be extended to multispectral and hyperspectral imagery. The basic approach of the algorithm is to extract a feature image from the RGB image that provides a good glare to background ratio, to detect the glare regions in the feature image, to extend the glare regions to cover all pixels that have been affected by the glare, and to remove the glare in the affected regions by filling in an estimate of the underlying image features. In our current implementation we use the green (G) image component as the feature image, given its high glare to background ratio and simplicity of calculation. Glare regions are either detected as saturated regions or small high contrasted bright regions. Saturated regions are detected using an adaptive thresholding method. Small high contrasted bright regions are detected using morphological top hat filters with different sizes and thresholds. The full extent of the glare regions is estimated by using a morphological constraint watershed segmentation to find the contour of the glare regions and adding a constant dilatation. The image features are estimated by interpolating the R,G,B color components individually from the surrounding regions based on Laplace's equation and modifying the intensity (I) component of the HSI color space transformed image. As the glare pattern is important to the physician, we embed it as a just visible intensity texture that does not affect the image processing. The performance of the algorithm is demonstrated using human subject data.

Methods for removing glare in digital endoscope images

“…Digital image processing methods have the advantage that they can operate on only selected parts of the image and thus offer greater versatility. Recent computational methods for removing specular glare, such as those mentioned earlier [8][9][10][11][12][13], have been designed for a range of surgery-related applications, which include processing that aims to restore the original tissue color to regions of glare. They are computationally intensive and not specifically intended for real-time applications.…”

“…A variety of inpainting methods have been reported for image quantification [8,9] and image registration (e.g., the beating heart [10]) and for single-image preprocessing before analysis of morphologically altered cervical tissue [11]. A technique known as light-field generation has been developed for image sequences.…”

Framework for hyperspectral image processing and quantification for cancer detection during animal tumor surgery

“…Optical images such as those from an endoscope and colposcopy images acquired during surgery are often strongly affected by glare spots in the images, which presents a major problem for surgical image analysis. 24 In HSI, glare alters the intensities of the pixels in each image band and consequently changes the spectral fingerprint, which could introduce artifacts in feature extraction and hence deteriorate classification results. Therefore, glare pixels need to be detected and removed before feature extraction and classification.…”

“…With high dimensional datasets, it is not a trivial task to extract the most relevant information from the raw data and classify it into tumor and nontumor tissue. Automated intraoperative cancer detection is especially challenging due to the motion artifacts and glares caused by specular reflection 24 and the high dimensionality of the dataset. 15 Therefore, we proposed a comprehensive workflow, as shown in Fig 1, which includes a set of preprocessing techniques as well as feature extraction and selection methods for intraoperative hyperspectral image analysis.…”

Synthesising Wider Field Images from Narrow-Field Retinal Video Acquired Using a Low-Cost Direct Ophthalmoscope (Arclight) Attached to a Smartphone

“…Our approach to detecting glare is inspired by the work of Lange [22] on images of the uterine cervix. Broadly speaking, we detect glare areas as those regions of the image which exhibit high contrast or saturation, and are approximately white.…”