Document Shadow Removal

14 papers with code • 0 benchmarks • 3 datasets

Document shadow removal refers to the process of eliminating or reducing the appearance of shadows in scanned or photographed documents. Shadows can occur due to various factors, such as uneven lighting, folds in the paper, or the presence of objects casting shadows during the scanning or capturing process.

Removing shadows from documents is important because they can degrade the readability and quality of the content. Shadows can obscure text or graphics, making it difficult to extract accurate information from the document. By eliminating shadows, the document becomes more legible and suitable for various purposes, including optical character recognition (OCR), document analysis, and archival purposes.

Document shadow removal techniques typically involve image processing and enhancement algorithms. These algorithms analyze the image and identify regions that contain shadows. They then adjust the brightness, contrast, and other image properties in the shadowed areas to minimize or eliminate the shadow effect. This process often requires advanced image analysis and manipulation techniques, such as histogram equalization, adaptive filtering, and image segmentation.

In recent years, machine learning and deep learning approaches have also been applied to document shadow removal. These methods utilize large datasets of shadowed and non-shadowed documents to train models that can automatically detect and remove shadows from new images. The models learn to recognize the characteristics of shadows and generate shadow-free versions of the documents.

Overall, document shadow removal plays a crucial role in improving the quality and legibility of scanned or photographed documents, making them more suitable for various applications in areas such as digital archiving, document analysis, and information extraction.

Most implemented papers

Removing Shadows from Images of Documents

tigerdeeda/Removing-Shadows-from-Images-of-Documents-by-Steve-Bako ACCV 2017

In this work, we automatically detect and remove distracting shadows from photographs of documents and other text-based items.

Water-Filling: An Efficient Algorithm for Digitized Document Shadow Removal

seungjun45/Water-Filling 22 Apr 2019

In this paper, we propose a novel algorithm to rectify illumination of the digitized documents by eliminating shading artifacts.

BEDSR-Net: A Deep Shadow Removal Network From a Single Document Image

IsHYuhi/BEDSR-Net_A_Deep_Shadow_Removal_Network_from_a_Single_Document_Image CVPR 2020

For taking advantage of specific properties of document images, a background estimation module is designed for extracting the global background color of the document.

High-Resolution Document Shadow Removal via A Large-Scale Real-World Dataset and A Frequency-Aware Shadow Erasing Net

CXH-Research/DocShadow-SD7K ICCV 2023

We handle high-resolution document shadow removal directly via a larger-scale real-world dataset and a carefully designed frequency-aware network.

Document Enhancement Using Visibility Detection

xuhangc/Document-Enhancement-using-Visibility-Detection CVPR 2018

Interestingly, this information is based on a solution to a seemingly unrelated problem of visibility detection in R3.

An Iterative Approach for Shadow Removal in Document Images

CV-Reimplementation/TraditionalDocumentShadowRemoval ICASSP 2018

Uneven illumination and shadows in document images cause a challenge for digitization applications and automated workflows.

An Effective Background Estimation Method for Shadows Removal of Document Images

CV-Reimplementation/TraditionalDocumentShadowRemoval ICIP 2019

This paper proposes an effective method to remove shadows from the single document images, which contains two stages: shadow detection and shadow removal.

Intrinsic Decomposition of Document Images In-the-Wild

cvlab-stonybrook/DocIIW 29 Nov 2020

However, document shadow or shading removal results still suffer because: (a) prior methods rely on uniformity of local color statistics, which limit their application on real-scenarios with complex document shapes and textures and; (b) synthetic or hybrid datasets with non-realistic, simulated lighting conditions are used to train the models.

Local Water-Filling Algorithm for Shadow Detection and Removal of Document Images

BingshuCV/DocumentShadowRemoval Sensors 2020

The proposed method can remove the shading artifacts and outperform some state-of-the-art methods, especially for the removal of shadow boundaries.

Document Shadow Removal with Foreground Detection Learning From Fully Synthetic Images

IsHYuhi/DSRFGD 2022 2022

In this paper, we present a large-scale and diverse dataset called fully synthetic document shadow removal dataset (FSDSRD) that does not require capturing documents.