In this paper we propose, a novel adaptative histogram matching method to remove low contrast, smeared ink, bleed-through and uneven illumination artefacts from scanned images of historical documents. The goal is to p...
详细信息
ISBN:
(纸本)9781728188577
In this paper we propose, a novel adaptative histogram matching method to remove low contrast, smeared ink, bleed-through and uneven illumination artefacts from scanned images of historical documents. The goal is to provide a better representation of document images and therefore improve readability and the source images for Optical Character recognition (OCR). Unlike other methods that are designed for single artefacts, our proposed method enhances multiple (low-contrast, smeared-ink, bleed-through and uneven illumination). The method starts by taking the bimodal peaks of the original grayscale image and multiplying them to generated gaussian windows to create an ideal histogram with weights of importance to distribution. This histogram becomes the reference histogram to be matched to the original image for a more optimized image. Median filtering is also incorporated in the method to remove salt and pepper noise. We demonstrate the technique on the European Newspapers project (ENP) dataset chosen from the patternrecognition and imageanalysis research lab (PRImA) and establish from the results that, the proposed method significantly reduces background noise and improves image quality on multiple artefacts as compared to other enhancement methods tested. To evaluate the efficiency of the proposed method, we make use of several performance criteria. These include Signal to Noise Ratio (SNR), Mean opinion score (MOS), and visual document image quality assessment (VDIQA) metric. The proposed method performs best in all the evaluation metrics having a 42.6 % increment on the average score of the other methods for MOS, 44.3% increment on average score of other methods for SNR and 61.11% better in VDIQA compared to other methods.
Haze during the bad weather, degrades the visibility of the scene drastically. Degradation of scene visibility varies with respect to the transmission coefficient/map (Tc) of the scene. Estimation of accurate Tc is ke...
详细信息
ISBN:
(纸本)9781450366151
Haze during the bad weather, degrades the visibility of the scene drastically. Degradation of scene visibility varies with respect to the transmission coefficient/map (Tc) of the scene. Estimation of accurate Tc is key step to reconstruct the haze free scene. Previously, local as well as global priors were proposed to estimate the Tc. We, on the other hand, propose integration of local and global approaches to learn both point level and object level Tc. The proposed local encoder decoder network (LEDNet) estimates the scene transmission map in two stages. During first stage, network estimates the point level Tc using parallel convolutional filters and spatial invariance filtering. The second stage comprises of a two level encoder-decoder architecture which anticipates the object level Tc. We also propose, local air-light estimation (LAE) algorithm, which is able to obtain the air-light component of the outdoor scene. Combination of LEDNet and LAE improves the accuracy of haze model to recover the scene radiance. Structural similarity index, mean square error and peak signal to noise ratio are used to evaluate the performance of the proposed approach for single image haze removal. Experiments on benchmark datasets show that LEDNet outperforms the existing state-of-the-art methods for single image haze removal.
In this work, we introduce a new architectural component to Neural Network (NN), i.e., trainable and spectrally initializable matrix transformations on feature maps. While previous literature has already demonstrated ...
详细信息
In this work, we introduce a new architectural component to Neural Network (NN), i.e., trainable and spectrally initializable matrix transformations on feature maps. While previous literature has already demonstrated the possibility of adding static spectral transformations as feature processors, our focus is on more general trainable transforms. We study the transforms in various architectural configurations on four datasets of different nature: from medical (ColorectalHist, HAM10000) and natural (Flowers) images to historical documents (CB55). With rigorous experiments that control for the number of parameters and randomness, we show that networks utilizing the introduced matrix transformations outperform vanilla neural networks. The observed accuracy increases appreciably across all datasets. In addition, we show that the benefit of spectral initialization leads to significantly faster convergence, as opposed to randomly initialized matrix transformations. The transformations are implemented as auto-differentiable PyTorch modules that can be incorporated into any neural network architecture. The entire code base is open-source.
This paper presents an objective comparative evaluation of page analysis and recognition methods for historical documents with text mainly in Bengali language and script. It describes the competition rules, dataset, a...
详细信息
Pathologist-defined labels are the gold standard for histopathological data sets, regardless of well-known limitations in consistency for some tasks. To date, some datasets on mitotic figures are available and were us...
详细信息
Since Arabic writing has a robust baseline, several state-of-the-art recognition systems for handwritten Arabic produce use of baseline-dependent characteristics. For modem Arabic documents, the baseline can be detect...
详细信息
ISBN:
(数字)9781839531088
Since Arabic writing has a robust baseline, several state-of-the-art recognition systems for handwritten Arabic produce use of baseline-dependent characteristics. For modem Arabic documents, the baseline can be detected reliably by obtaining the maximum in the horizontal projection profile or the Hough transformed image. However, the performance of these techniques leaks significantly on Historical Arabic Documents. In this paper, we introduce an effective novel approach to baseline detection in Historical Arabic Documents which is based on Based on Voronoi Diagrams. The proposed technique is carried out, verified and validated on a dataset of Warped Historical Arabic Documents based on affecting by warping percentage.
The COVID-19 pandemic exposed a global deficiency of systematic, data-driven guidance to identify high-risk individuals. Here, we illustrate the utility of routinely recorded medical history to predict the risk for 17...
The COVID-19 pandemic exposed a global deficiency of systematic, data-driven guidance to identify high-risk individuals. Here, we illustrate the utility of routinely recorded medical history to predict the risk for 1741 diseases across clinical specialties and support the rapid response to emerging health threats such as COVID-19. We developed a neural network to learn from health records of 502,489 UK Biobank participants. Importantly, we observed discriminative improvements over basic demographic predictors for 1546 (88.8%) endpoints. After transferring the unmodified risk models to the All of US cohort, we replicated these improvements for 1115 (78.9%) of 1414 investigated endpoints, demonstrating generalizability across healthcare systems and historically underrepresented groups. Ultimately, we showed how this approach could have been used to identify individuals vulnerable to severe COVID-19. Our study demonstrates the potential of medical history to support guidance for emerging pandemics by systematically estimating risk for thousands of diseases at once at minimal cost.
This paper introduces a new way for text-line extraction by integrating deep-learning based pre-classification and state-of-the-art segmentation methods. Text-line extraction in complex handwritten documents poses a s...
详细信息
This paper introduces a new way for text-line extraction by integrating deep-learning based pre-classification and state-of-the-art segmentation methods. Text-line extraction in complex handwritten documents poses a significant challenge, even to the most modern computer vision algorithms. Historical manuscripts are a particularly hard class of documents as they present several forms of noise, such as degradation, bleed-through, interlinear glosses, and elaborated scripts. In this work, we propose a novel method which uses semantic segmentation at pixel level as intermediate task, followed by a text-line extraction step. We measured the performance of our method on a recent dataset of challenging medieval manuscripts and surpassed state-of-the-art results by reducing the error by 80.7%. Furthermore, we demonstrate the effectiveness of our approach on various other datasets written in different scripts. Hence, our contribution is two-fold. First, we demonstrate that semantic pixel segmentation can be used as strong denoising pre-processing step before performing text line extraction. Second, we introduce a novel, simple and robust algorithm that leverages the high-quality semantic segmentation to achieve a text-line extraction performance of 99.42% line IU on a challenging dataset.
The density of mitotic figures within tumor tissue is known to be highly correlated with tumor proliferation and thus is an important marker in tumor grading. recognition of mitotic figures by pathologists is known to...
详细信息
暂无评论