Is learning by reading a book better than watching a movie? A computational analysis of semantic concept network growth during text and multimedia comprehension

Author(s):  
Naser S Al Madi ◽  
Javed I. Khan
2021 ◽  
Author(s):  
Thomas Hegghammer

Optical Character Recognition (OCR) can open up understudied historical documents to computational analysis, but the accuracy of OCR software varies. This article reports a benchmarking experiment comparing the performance of Tesseract, Amazon Textract, and Google Document AI on images of English and Arabic text. English-language book scans (n=322) and Arabic-language article scans (n=100) were replicated 43 times with different types of artificial noise for a corpus of 18,568 documents, generating 51,304 process requests. Document AI delivered the best results, and the server-based processors (Textract and Document AI) were substantially more accurate than Tesseract, especially on noisy documents. Accuracy for English was considerably better than for Arabic. Specifying the relative performance of three leading OCR products and the differential effects of commonly found noise types can help scholars identify better OCR solutions for their research needs. The test materials have been preserved in the openly available "Noisy OCR Dataset" (NOD).


Author(s):  
Thomas Hegghammer

AbstractOptical Character Recognition (OCR) can open up understudied historical documents to computational analysis, but the accuracy of OCR software varies. This article reports a benchmarking experiment comparing the performance of Tesseract, Amazon Textract, and Google Document AI on images of English and Arabic text. English-language book scans (n = 322) and Arabic-language article scans (n = 100) were replicated 43 times with different types of artificial noise for a corpus of 18,568 documents, generating 51,304 process requests. Document AI delivered the best results, and the server-based processors (Textract and Document AI) performed substantially better than Tesseract, especially on noisy documents. Accuracy for English was considerably higher than for Arabic. Specifying the relative performance of three leading OCR products and the differential effects of commonly found noise types can help scholars identify better OCR solutions for their research needs. The test materials have been preserved in the openly available “Noisy OCR Dataset” (NOD) for reuse in future benchmarking studies.


2018 ◽  
Vol 28 (2) ◽  
pp. 153-169 ◽  
Author(s):  
Kayo Gonçalves-E-Silva ◽  
Daniel Aloise ◽  
Samuel Xavier-De-Souza ◽  
Nenad Mladenovic

Nelder-Mead method (NM) for solving continuous non-linear optimization problem is probably the most cited and the most used method in the optimization literature and in practical applications, too. It belongs to the direct search methods, those which do not use the first and the second order derivatives. The popularity of NM is based on its simplicity. In this paper we propose even more simple algorithm for larger instances that follows NM idea. We call it Simplified NM (SNM): instead of generating all n + 1 simplex points in Rn, we perform search using just q + 1 vertices, where q is usually much smaller than n. Though the results cannot be better than after performing calculations in n+1 points as in NM, significant speed-up allows to run many times SNM from different starting solutions, usually getting better results than those obtained by NM within the same cpu time. Computational analysis is performed on 10 classical convex and non-convex instances, where the number of variables n can be arbitrarily large. The obtained results show that SNM is more effective than the original NM, confirming that LIMA yields good results when solving a continuous optimization problem.


2019 ◽  
Vol 36 (05) ◽  
pp. 1950030 ◽  
Author(s):  
Tao Du ◽  
Danyang Li ◽  
Yan Ren ◽  
Chong Lu ◽  
Wanquan Liu

In this paper, a revised directional triangle-area curve representation method (DTAR) is proposed to address the problem of eyebrow semantic shape characterization via curve representation. First, 11 or 12 DTAR values are selected to describe eyebrows via considering the eyebrow corner information roughly, and then the corresponding DTAR curves are acquired via the cubic spline interpolation based on these selected points. Second, a descriptor of the landmarks is developed to represent selected reference eyebrows, and the corresponding DTAR curves are obtained for the selected reference eyebrows. Lastly, a similarity notion based on AFS is introduced via measuring the membership degrees of each eyebrow shape similar to the given reference shapes, and then one can describe each eyebrow shape by using two given reference eyebrow shapes via computing the membership degrees representing the relative similarities. To illustrate the effectiveness of the proposed approach, we use the AR and BJUT databases for experiments to demonstrate the consistency comparison with human perceptions. The experimental results show that the extracted semantic notions of eyebrow shapes obtained by the proposed approach are much better than those by only utilizing 11 DTAR values or 12 DTAR values directly in terms of the consistency with human perceptions.


1972 ◽  
Vol 1 ◽  
pp. 27-38
Author(s):  
J. Hers

In South Africa the modern outlook towards time may be said to have started in 1948. Both the two major observatories, The Royal Observatory in Cape Town and the Union Observatory (now known as the Republic Observatory) in Johannesburg had, of course, been involved in the astronomical determination of time almost from their inception, and the Johannesburg Observatory has been responsible for the official time of South Africa since 1908. However the pendulum clocks then in use could not be relied on to provide an accuracy better than about 1/10 second, which was of the same order as that of the astronomical observations. It is doubtful if much use was made of even this limited accuracy outside the two observatories, and although there may – occasionally have been a demand for more accurate time, it was certainly not voiced.


Author(s):  
J. Frank ◽  
P.-Y. Sizaret ◽  
A. Verschoor ◽  
J. Lamy

The accuracy with which the attachment site of immunolabels bound to macromolecules may be localized in electron microscopic images can be considerably improved by using single particle averaging. The example studied in this work showed that the accuracy may be better than the resolution limit imposed by negative staining (∽2nm).The structure used for this demonstration was a halfmolecule of Limulus polyphemus (LP) hemocyanin, consisting of 24 subunits grouped into four hexamers. The top view of this structure was previously studied by image averaging and correspondence analysis. It was found to vary according to the flip or flop position of the molecule, and to the stain imbalance between diagonally opposed hexamers (“rocking effect”). These findings have recently been incorporated into a model of the full 8 × 6 molecule.LP hemocyanin contains eight different polypeptides, and antibodies specific for one, LP II, were used. Uranyl acetate was used as stain. A total of 58 molecule images (29 unlabelled, 29 labelled with antl-LPII Fab) showing the top view were digitized in the microdensitometer with a sampling distance of 50μ corresponding to 6.25nm.


Author(s):  
A. V. Crewe

We have become accustomed to differentiating between the scanning microscope and the conventional transmission microscope according to the resolving power which the two instruments offer. The conventional microscope is capable of a point resolution of a few angstroms and line resolutions of periodic objects of about 1Å. On the other hand, the scanning microscope, in its normal form, is not ordinarily capable of a point resolution better than 100Å. Upon examining reasons for the 100Å limitation, it becomes clear that this is based more on tradition than reason, and in particular, it is a condition imposed upon the microscope by adherence to thermal sources of electrons.


Sign in / Sign up

Export Citation Format

Share Document