Document image enhancement methods are often used to improve the accuracy and efficiency of automated document analysis and recognition tasks such as character recognition. These document images ...could be degraded or damaged for various reasons including aging, fading handwriting, poor lighting conditions, watermarks, etc. In recent years, with the improvement of computer performance and the continuous development of deep learning, many methods have been proposed to enhance the quality of these document images. In this paper, we review six tasks of document degradation, namely, background texture, page smudging, fading, poor lighting conditions, watermarking, and blurring. We summarize the main models for each degradation problem as well as recent work, such as the binarization model that can be used to deal with the degradation of background textures, lettering smudges. When facing the problem of fading, a model for stroke connectivity can be used, while the other three degradation problems are mostly deep learning models. We discuss the current limitations and challenges of each degradation task and introduce the common public datasets and metrics. We identify several promising research directions and opportunities for future research.
Abstract
This article presents a novel functional verification methodology based on the Universal Verification Methodology (UVM) to validate the functionality of the Quick Serial Peripheral Interface ...(QSPI) module. QSPI serves as a prevalent serial communication protocol widely employed for high-speed data exchange with external flash devices. By using the code coverage and functional coverage reports provided by UVM, we were able to assess how well the test cases covered the design code, identifying the parts of the design that had been adequately tested and those that needed more test coverage. This verification method based on UVM provides powerful tools and technical support for the design and development of the QSPI module and effectively ensures its reliability and stability. This study is of great significance for further improving the design quality and verification efficiency of the QSPI module.
Abstract
A metamaterial absorber consisting of stacked and orthogonal elliptical graphene layers is proposed to achieve eight absorption bands in the terahertz (THz) frequency range. By adjusting the ...length of the short or long axis for each graphene ellipse, the corresponding absorption peak can be tuned independently. Electric field distributions are present to reveal the physical mechanism. The absorption spectrum can also be tuned by varying the chemical potential of each graphene ellipse and the manipulation for the polarization angle of the incidence. Moreover, the absorber can work properly under oblique incidences. Our work paves a way for designing tunable THz absorbers with numerous absorption bands.
Document image binarization is a challenging task, especially when it comes to text segmentation in degraded document images. The binarization, as a pre-processing step of Optical Character ...Recognition (OCR), is one of the most fundamental and commonly used segmentation methods. It separates the foreground text from the background of the document image to facilitate subsequent image processing. In view of the different degradation degrees of document images, researchers have proposed a variety of solutions. In this paper, we have summarized some challenges and difficulties in the field of document image binarization. Approximately 60 methods documenting image binarization techniques are mentioned, including traditional algorithms and deep learning-based algorithms. Here, we evaluated the performance of 25 image binarization techniques on the H-DIBCO2016 dataset to provide some help for future research.
We theoretically investigate the anisotropic plasmonic resonances in the proposed infrared absorber, which consists of stacked graphene-black phosphorus (BP) bilayers with dual absorption peaks. By ...combining the advantages of graphene and BP, stacked graphene-black phosphorus bilayers exhibit high absorption rates at both peaks and strong anisotropy. The loss mechanism is revealed deeply with electric field distributions, while the near field coupling between graphene and BP is discussed detailedly. Furthermore, by altering the corresponding doping levels of graphene and BP, each of the absorption bands can be independently tuned effectively. The angular dependence for oblique incidence is illustrated by performing a series of simulations. Besides, polarization-sensitivity for stacked graphene-BP bilayers (GBPBs) is also presented. Thus, our approach provides a theoretical and systematic guide for designing a variety of multi-resonant graphene-BP-based spatial absorbers, which show potentials in the applications of sensors and reflective polarizers.
Background. With the highest mortality and metastasis rate, kidney renal clear cell carcinoma (KIRC) is one of the most common urological malignant tumors and not sensitive to chemotherapy and ...radiotherapy. Immunotherapy, which proves to be effective and a big progression, such as PD-1/PD-L1 inhibitors, is not sensitive to all KIRC patients. To predict prognosis and immunotherapy response, a novel immune checkpoint gene- (ICG-) related model is essential in clinics. Methods. From the public database-downloaded dataset, a novel ICG-related model for predicting prognosis and immunotherapy response in KIRC patients was built up and verified with R packages and Cox regression analysis. The Kaplan-Meier curve was plotted. Results. 39 ICGs were identified to have different expression in KIRC patients and enriched in immune-related biological pathways and activities. Three ICGs (CTLA4, TNFSF14, and HHLA2) were screened to generate KIRC-ICG model. The KIRC-ICG model was verified to be effective. With conducting KIRC-SYS model, KIRC-ICGscore was verified to be an independent factor regardless of age, gender, stage, grade, and TNM stage. Compared to the ICG-low subgroup, the ICG-high subgroup had more immune activities. KIRC-ICGscore was significantly positively correlated with the expression of Treg markers. KIRC-ICG model could also be reliable to predict immunotherapy response. Conclusion. The KIRC-ICG model was reliable to predict prognosis and immunotherapy response for KIRC patients and could be an independent factor regardless of clinical characteristics.
A Cache Hardware Design for H.264 Encoder Shikai Zuo; Mingjiang Wang; Liyi Xiao
2012 Second International Conference on Instrumentation, Measurement, Computer, Communication and Control,
2012-Dec.
Conference Proceeding
On the oretical basis of the inter-frame motion compensation of H.264 encoder, this paper proposed a new cache which adopted the LFU(lease frequently used) replace mechanism and a new prefetching ...method of reference images which have been reconstructed in H.264 encoder. We first gave its top-level framework definition, and then implemented the design using Verilog HDL. Based on the RTL coding, the design is synthesized with TSMC 90nm technology library. The result of the related experiments shows that the proposed cache can effectively save 76%~86% of the data memory bandwidth by use of the temporal correlation of data. Compared with the traditional method, the new cache architecture shows much better performance on reducing the bandwidth of accessing main storage, and presents much lower power consumption, higher hit ratio and faster access speed.
The increasing functional requirements of IC circuits make the verification stimulus complexity exponentially increasing, so the SystemVerilog language based UVM general verification methodology is ...gradually becoming the main verification method. The verification methodology based on SystemVerilog language UVM with verification methodology is used to design and verify the acceptance filtering module of CAN. The verification platform uses SystemVerilog to generate the UVM framework structure using Python automation scripts, combined with constrainable random testing techniques to write multiple test cases for functional points. The verification simulation results show that the verification coverage reaches 100%. In addition, the verification platform is easy to migrate, which can greatly improve the verification efficiency and shorten the verification time.