X-ray imaging is one of the most widely used security measures for maintaining airport and transportation security. Conventional X-ray imaging systems typically apply tone-mapping (TM) algorithms to visualize high-dynamic-range (HDR) X-ray images on a standard 8-bit display device. However, X-ray images obtained through traditional TM algorithms often suffer from halo artifacts or detail loss in inter-object overlapping regions, which makes it difficult for an inspector to detect unsafe or hazardous objects. To alleviate these problems, this article proposes a deep learning-based TM method for X-ray inspection. The proposed method consists of two networks called detail-recovery network (DR-Net) and TM network (TM-Net). The goal of DR-Net is to restore the details in the input HDR image, whereas TM-Net aims to compress the dynamic range while preserving the restored details and preventing halo artifacts. Since there are no standard ground-truth images available for the TM of X-ray images, we propose a novel loss function for unsupervised learning of TM-Net. We also introduce a dataset synthesis technique using the Beer-Lambert law for supervised learning of DR-Net. Extensive experiments comparing the performance of our proposed method with state-of-the-art TM methods demonstrate that the proposed method not only achieves visually compelling results but also improves the quantitative performance measures such as FSITM and HDR-VDP-2.2.

/pdf/detail-restoration-and-tone-mapping-networks-for-x-ray-2x7qv7roqn.pdf

Detail Restoration and Tone Mapping Networks for X-Ray Security Inspection

Medical imaging can help doctors in better diagnosis of several conditions. During the present COVID-19 pandemic, timely detection of novel coronavirus is crucial, which can help in curing the disease at an early stage. Image enhancement techniques can improve the visual appearance of COVID-19 CT scans and speed-up the process of diagnosis. In this study, we analyze some state-of-the-art image enhancement techniques for their suitability in enhancing the CT scans of COVID-19 patients. Six quantitative metrics, Entropy, SSIM, AMBE, PSNR, EME, and EMEE, are used to evaluate the enhanced images. Two experienced radiologists were involved in the study to evaluate the performance of the enhancement techniques and the quantitative metrics used to assess them. 

/pdf/a-comparative-study-of-medical-image-enhancement-algorithms-24x20xji.pdf

A comparative study of medical image enhancement algorithms and quality assessment metrics on COVID-19 CT images

This paper introduces a new learning based framework for X-ray images that relies on a morphological decomposition of the signal into two main components, separating images into local textures and piecewise smooth (cartoon) parts. The piecewise smooth component corresponds to the spatial variation of the average density of the objects, whereas the local texture component presents the inspected objects singularities. Our method builds on two convolutional neural network (CNN) branches to decompose an input image into its two morphological components. This CNN is trained with synthetic data, generated by randomly picking piecewise smooth and singular patterns in a parametric dictionary and enforcing the sum of the CNN branches to approximate the identity mapping. We demonstrate the relevance of the decomposition by enhancing the local textures component compared to the piecewise smooth one. Those enhanced images compare favorably to the ones obtained with existing works destined to visualize High Dynamic Range (HDR) images such as tone-mapping algorithms.

/pdf/cnn-based-morphological-decomposition-of-x-ray-images-for-1z33wc3jnh.pdf

CNN-based morphological decomposition of X-ray images for details and defects contrast enhancement

For the problems of darkness, insufficient contrast, and color cast in the images captured by CMOS image sensors in the <inline-formula> <tex-math notation="LaTeX">$\gamma $ </tex-math></inline-formula> radiation environment, this paper proposed a joint contrast improvement and color cast correction approach for the <inline-formula> <tex-math notation="LaTeX">$\gamma $ </tex-math></inline-formula> radiation image enhancement, which improved the image effects from enhancing detail expression and color expression. For the problem of overall darkness of <inline-formula> <tex-math notation="LaTeX">$\gamma $ </tex-math></inline-formula> radiation image, we first utilized logarithmic mapping to improve the image brightness. Secondly, for the insufficient contrast of image details, a contrast enhancement method with an adaptive Gamma coefficient is proposed, which adaptively adjusts the image brightness at the pixel level, so that the image detail expression is more in line with the human visual characteristics (HVC). Lastly, the color level remapping method is exploited for color cast correction. Extensive experiments are conducted on the <inline-formula> <tex-math notation="LaTeX">$\gamma $ </tex-math></inline-formula> radiation images collected in the Co60 environment. With the help of the proposed method, we can achieve the best results in quantitative and visual comparison. Experimental results demonstrate that the proposed method enjoys state-of-the-art performance in <inline-formula> <tex-math notation="LaTeX">$\gamma $ </tex-math></inline-formula> radiation image enhancement.

γ Radiation Image Enhancement Method Based on Non-Linear Mapping

Radiographic testing is extensively used in non-destructive evaluation. However, the contrast of object defects in X-Ray images is generally weak compared to the image dynamic range, which limits defects visibility and makes the image inspection time consuming and tedious. To circumvent this issue, our work builds on a bilateral filter to extract and amplify the low-contrast local variations of the image, supposed to represent the object defects. A critical step of this approach lies in the selection of the bilateral filter range parameter, which sets a contrast upper-bound to the signals that are filtered out, to be scaled to a constant fraction of the rendering range. As an original contribution, we propose to set this parameter to maximize the tone-mapped quality index (TMQI) of the reconstructed image. Since the TMQI combines a multi-scale structural similarity criterion with a measure of image naturalness, its maximization both discourages the selection of a (too large) range parameter that penalizes structural similarity, and prevents the lack of naturalness associated to an excessive noise amplification caused by a too small range parameter. Image subjective quality assessment experiments not only demonstrate that our approach largely enhances defect visibility on real use cases, but also reveal that our recommended parameter is preferred by users to any reasonable alternative in a large majority of cases ( 90%). An objective evaluation based on the Image Discrete Entropy shows that the enhanced images compare favorably to the ones obtained with previous works devoted to High Dynamic Range image visualization

Optimizing the range of a bilateral filter to enhance defects and threads visibility in x-ray radiographic testing

This paper introduces a novel tone mapping operator, designed to offer a good rendering of the local structures. The new operator fusions the multiple versions of a single HDR input obtained by clipping and normalizing its intensity based on a complete set of disjoint intervals. Defining the weight map associated to each version to be its clipping interval indicator function promotes contrast enhancement, but induces artifacts when neighboring pixels belong to distinct intervals. We thus propose to smooth out the indicators across neighboring pixels with similar intensity, using a standard cross-bilateral filter. With such weight maps, the fusion operator becomes equivalent to applying histogram equalization on the image regions on which the cross-bilateral filter diffuses the indicators, and is therefore referred to as Bilateral Histogram Equalization (BHE) operator. It compares favorably to previous tone mapping algorithms.

Tahani Madmad

Papers

Bilateral Histogram Equalization for X-Ray Image Tone Mapping

CNN-based morphological decomposition of X-ray images for details and defects contrast enhancement

Optimizing the range of a bilateral filter to enhance defects and threads visibility in x-ray radiographic testing