scanned image
Recently Published Documents


TOTAL DOCUMENTS

103
(FIVE YEARS 35)

H-INDEX

7
(FIVE YEARS 2)

Author(s):  
Shilpa Pandey ◽  
Gaurav Harit

In this article, we address the problem of localizing text and symbolic annotations on the scanned image of a printed document. Previous approaches have considered the task of annotation extraction as binary classification into printed and handwritten text. In this work, we further subcategorize the annotations as underlines, encirclements, inline text, and marginal text. We have collected a new dataset of 300 documents constituting all classes of annotations marked around or in-between printed text. Using the dataset as a benchmark, we report the results of two saliency formulations—CRF Saliency and Discriminant Saliency, for predicting salient patches, which can correspond to different types of annotations. We also compare our work with recent semantic segmentation techniques using deep models. Our analysis shows that Discriminant Saliency can be considered as the preferred approach for fast localization of patches containing different types of annotations. The saliency models were learned on a small dataset, but still, give comparable performance to the deep networks for pixel-level semantic segmentation. We show that saliency-based methods give better outcomes with limited annotated data compared to more sophisticated segmentation techniques that require a large training set to learn the model.


2021 ◽  
Vol 12 (4-2021) ◽  
pp. 154-161
Author(s):  
I. A. Travin ◽  

The article considers the issue of the researcher's perception of the scanned image of paper sheets: both separate and as part of a book. The importance and timeliness of work on obtaining digital copies of books and sheets with text / photographs was emphasized. The problem is the authenticity of their depiction of a physical object. The methods of scanning and visual features of the image on an electronic screen are characterized, depending on whether the work is carried out to scan a paper sheet in its entirety or minus the margins and edges of the sheet. Currently, there are no technologies for transferring the texture of a paper sheet when scanning, which leads to an erroneous solution to this problem by increasing the clarity of scanning. The greatest authenticity of the image of a physical object can be achieved by scanning the entire sheet, without deliberately separating the margins and edges of the sheet.


Entropy ◽  
2021 ◽  
Vol 23 (10) ◽  
pp. 1259
Author(s):  
Joao Florindo ◽  
Konradin Metze

Here we present a study on the use of non-additive entropy to improve the performance of convolutional neural networks for texture description. More precisely, we introduce the use of a local transform that associates each pixel with a measure of local entropy and use such alternative representation as the input to a pretrained convolutional network that performs feature extraction. We compare the performance of our approach in texture recognition over well-established benchmark databases and on a practical task of identifying Brazilian plant species based on the scanned image of the leaf surface. In both cases, our method achieved interesting performance, outperforming several methods from the state-of-the-art in texture analysis. Among the interesting results we have an accuracy of 84.4% in the classification of KTH-TIPS-2b database and 77.7% in FMD. In the identification of plant species we also achieve a promising accuracy of 88.5%. Considering the challenges posed by these tasks and results of other approaches in the literature, our method managed to demonstrate the potential of computing deep learning features over an entropy representation.


2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Ain Ashraf Rizwal ◽  
Nursyereen Azahar ◽  
Nor Hidayah Reduwan ◽  
Mohd Yusmiaidil Putera Mohd Yusof

Abstract Background Preservation of bite marks evidence has always been a major problem in forensic odontology due to progressive loss of details as time passes. The use of 2D photographs has been widely used to document forensic evidence and preserving bite marks; however, there are limitations to this method. This study aims to measure the accuracy of the 3D scanned image in comparison to 2D photograph registration of experimental bite marks. Thirty volunteers performed self-exertions of a bite mark on the respective forearm of subjects. A 2D photograph and 3D scanned image was immediately registered following bite mark exercise using a conventional camera and Afinia EinScan-Pro 2X PLUS Handheld 3D Scanner, respectively. The outlines of the bite mark were transformed into a polygonal shape. Next, the polygonal approximation analysis was performed by an arbitrary superimposition method. The difference between surface areas of both images was calculated (2D photographs ̶ 3D scanned images). Results A paired t test was used to measure significance with α = 0.05. The mean surface area of 2D photographs and 3D scanned images is 31.535 cm2 and 31.822 cm2, respectively. No statistical difference was found between both mean surface areas (p > 0.05). The mean error (ME) is 0.287 ± 3.424 cm2 and the mean absolute error (MAE) is 1.733 ± 1.149 cm2. Conclusion Bite marks registered with the 3D scanned image are comparable to the standard 2D photograph for bite mark evaluations. The use of a 3D scan may be adopted as a standard operating procedure in the forensic application, especially for evidence preservation.


Author(s):  
Sushmitha M

Communication is the basic requirement for humans to connect and it requires text and speech but visually impaired people cannot able to perform this. This project helps them to read the image. This project is an automatic document reader for visually impaired people, developed on the Raspberry Pi processor board. It controls the peripherals like a camera, a speaker which acts as an interface between the system and the user. Here, we use a raspberry pi camera which is used to capture the image and scan the image using Image Magick software. Then the output of the scanned image is given to OCR(optical character recognition) software to convert the image to text. It converts the typed or printed text to the machine code. Then we use Text to Speech (TTS), which is used to convert speech to text. The experimental result is very helpful to blind people as there was much analysis of the different objects.


2021 ◽  
Vol 11 (11) ◽  
pp. 4894
Author(s):  
Anna Scius-Bertrand ◽  
Michael Jungo ◽  
Beat Wolf ◽  
Andreas Fischer ◽  
Marc Bui

The current state of the art for automatic transcription of historical manuscripts is typically limited by the requirement of human-annotated learning samples, which are are necessary to train specific machine learning models for specific languages and scripts. Transcription alignment is a simpler task that aims to find a correspondence between text in the scanned image and its existing Unicode counterpart, a correspondence which can then be used as training data. The alignment task can be approached with heuristic methods dedicated to certain types of manuscripts, or with weakly trained systems reducing the required amount of annotations. In this article, we propose a novel learning-based alignment method based on fully convolutional object detection that does not require any human annotation at all. Instead, the object detection system is initially trained on synthetic printed pages using a font and then adapted to the real manuscripts by means of self-training. On a dataset of historical Vietnamese handwriting, we demonstrate the feasibility of annotation-free alignment as well as the positive impact of self-training on the character detection accuracy, reaching a detection accuracy of 96.4% with a YOLOv5m model without using any human annotation.


Materials ◽  
2021 ◽  
Vol 14 (9) ◽  
pp. 2340
Author(s):  
Hyun-Su Oh ◽  
Young-Jun Lim ◽  
Bongju Kim ◽  
Myung-Joo Kim ◽  
Ho-Beom Kwon ◽  
...  

This study was performed to verify the influence of scanning-aid materials on the accuracy and time efficiency of full-arch scanning with intraoral scanners. The full-arch reference model was constructed by a 3D printer and scanned with a model scanner to obtain the reference dataset. Four experimental groups (application of ScanCure (SC-80, ODS Co, Incheon, Korea), IP Scan Spray (IP-Division, Haimhausen, Germany) and Vita Powder Scan Spray (Vita Zahnfabrik, Stuttgart, Germany), and no treatment) were designed, and the scans were executed (trueness, n = 5) using two intraoral scanners: I500 (Medit Co., Seoul, Korea) and TRIOS (3shape, Copenhagen, Denmark). All acquired scan data were compared with the reference datasets using the 3D superimposition method and 2D linear measurements. In the 3D analysis, intragroup data were compared with each other (precision, n = 10). Time efficiency was also verified by comparing the scan times of the four experimental groups. In the 3D analysis, the root mean square (RMS) value of the precision of the scanned image was statistically significantly more accurate in the scanning-aid agent-treated groups than in the no-treatment group (p < 0.05). However, the RMS values of trueness and the types of scanning-aid materials were not significantly different. In the 2D measurements, the increased scan distance generated a greater distance deviation. The working time was significantly shorter in the scanning-aid agent groups than in the no-treatment group, with statistical significance (p < 0.05). Therefore, in clinical situations, the application of scanning-aid materials is recommended to reduce scanning time and more efficiently obtain the full-arch scanned image.


Electronics ◽  
2021 ◽  
Vol 10 (3) ◽  
pp. 320
Author(s):  
Shundao Xie ◽  
Hong-Zhou Tan

Traceability is considered a promising solution for product safety. However, the data in the traceability system is only a claim rather than a fact. Therefore, the quality and safety of the product cannot be guaranteed since we cannot ensure the authenticity of products (aka counterfeit detection) in the real world. In this paper, we focus on counterfeit detection for the traceability system. The risk of counterfeiting throughout a typical product life cycle in the supply chain is analyzed, and the corresponding requirements for the tags, packages, and traceability system are given to eliminate these risks. Based on the analysis, an anti-counterfeiting architecture for traceability system based on two-level quick response codes (2LQR codes) is proposed, where the problem of counterfeit detection for a product is transformed into the problem of copy detection for the 2LQR code tag. According to the characteristics of the traceability system, the generation progress of the 2LQR code is modified, and there is a corresponding improved algorithm to estimate the actual location of patterns in the scanned image of the modified 2LQR code tag to improve the performance of copy detection. A prototype system based on the proposed architecture is implemented, where the consumers can perform traceability information queries by scanning the 2LQR code on the product package with any QR code reader. They can also scan the 2LQR code with a home-scanner or office-scanner, and send the scanned image to the system to perform counterfeit detection. Compared with other anti-counterfeiting solutions, the proposed architecture has advantages of low cost, generality, and good performance. Therefore, it is a promising solution to replace the existing anti-counterfeiting system.


2021 ◽  
Vol 29 (2) ◽  
pp. 552
Author(s):  
Benjamin S. Sajdak ◽  
Jack T. Postlewaite ◽  
Kevin W. Eliceiri ◽  
Jeremy D. Rogers

2021 ◽  
Vol 40 ◽  
pp. 03009
Author(s):  
Shristi Mittal ◽  
Rhutuja Satpute ◽  
Shubhamm Mohitte ◽  
Leena Ragha ◽  
Dhanashri Bhosale

Sketches are commonly used in the fields of engineering and architecture, especially for the early design phases. Engineers spend considerable time setting up initial designs using pencil and paper, and then redrawing them to any software. This problem can be solved by using the idea to scan the circuit sketch with android device which is drawn on the paper and translate it into standard layouts and run circuit simulations. The scanned image will be pre-processed and further segmented. The segmented image will be used to extract the features which are in turn given for classification. Recognizing sketches may seem so quick and intuitive to humans but it is really a big challenge for the machine. In this proposed work the aim is to achieve high precision trainable electronic circuit component recognizer for sketched circuits with fast response time and simple extensibility to new components.


Sign in / Sign up

Export Citation Format

Share Document