Confidence Prediction for Lexicon-Free OCR

Mor, Noam; Wolf, Lior

doi:10.1109/wacv.2018.00030

Cited by 18 publications

(10 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Note, however, that like the other rejection approaches, this carves out a small region around the origin, and potentially between two classes as "none of the above," and still leaves infinite open space risk and hence does not solve OSR. In addition, there is also active research in network uncertainty estimation (Gal and Ghahramani 2016;Lakshminarayanan, Pritzel, and Blundell 2017;Mor and Wolf 2018). The authors of such claim thresholding their uncertainity can reject outliers.…”

Section: Open Set Deep Networkmentioning

confidence: 99%

Learning and the Unknown: Surveying Steps toward Open World Recognition

Boult

Cruz

Dhamija

et al. 2019

AAAI

109

View full text Add to dashboard Cite

As science attempts to close the gap between man and machine by building systems capable of learning, we must embrace the importance of the unknown. The ability to differentiate between known and unknown can be considered a critical element of any intelligent self-learning system. The ability to reject uncertain inputs has a very long history in machine learning, as does including a background or garbage class to account for inputs that are not of interest. This paper explains why neither of these is genuinely sufficient for handling unknown inputs – uncertain is not unknown, and unknowns need not appear to be uncertain to a learning system. The past decade has seen the formalization and development of many open set algorithms, which provably bound the risk from unknown classes. We summarize the state of the art, core ideas, and results and explain why, despite the efforts to date, the current techniques are genuinely insufficient for handling unknown inputs, especially for deep networks.

show abstract

Section: Open Set Deep Networkmentioning

confidence: 99%

Learning and the Unknown: Surveying Steps toward Open World Recognition

Boult

Cruz

Dhamija

et al. 2019

AAAI

109

View full text Add to dashboard Cite

show abstract

“…The LSTM model. We implement the OCR model described in [22], which consists of a convolutional layer, followed by a 3-layer deep bidirectional LSTM, and optimizes for CTC loss. The trained model achieves a precision score of 85.7% on the test set of ORAND-CAR-A, which would have achieved first place in the HDSRC 2014 competition.…”

Section: Lstm On Handwritten Numbersmentioning

confidence: 99%

Adversarial Attacks on Binary Image Recognition Systems

Balkanski¹,

Chase²,

Oshiba³

et al. 2020

Preprint

View full text Add to dashboard Cite

We initiate the study of adversarial attacks on models for binary (i.e. black and white) image classification. Although there has been a great deal of work on attacking models for colored and grayscale images, little is known about attacks on models for binary images. Models trained to classify binary images are used in text recognition applications such as check processing, license plate recognition, invoice processing, and many others. In contrast to colored and grayscale images, the search space of attacks on binary images is extremely restricted and noise cannot be hidden with minor perturbations in each pixel. Thus, the optimization landscape of attacks on binary images introduces new fundamental challenges.In this paper we introduce a new attack algorithm called Scar, designed to fool classifiers of binary images. We show that Scar significantly outperforms existing L 0 attacks applied to the binary setting and use it to demonstrate the vulnerability of real-world text recognition systems. Scar's strong performance in practice contrasts with hardness results that show the existence of classifiers that are provably robust to large perturbations. In many cases, altering a single pixel is sufficient to trick Tesseract, a popular open-source text recognition system, to misclassify a word as a different word in the English dictionary. We also license software from providers of check processing systems to most of the major US banks and demonstrate the vulnerability of check recognitions for mobile deposits. These systems are substantially harder to fool since they classify both the handwritten amounts in digits and letters, independently. Nevertheless, we generalize Scar to design attacks that fool state-of-the-art check processing systems using unnoticeable perturbations that lead to misclassification of deposit amounts. Consequently, this is a powerful method to perform financial fraud.

show abstract

“…The prediction accuracy of the OCR engine such as Tesseract [9] poses a limitation to our toolkit. The accuracy of Tesseract OCR is not always 100% [14], [15]. To handle this limitation and to enhance the accuracy, we used regionbased segmentation and image preprocessing techniques such as noise removal, canny edge detection, and contours finding.…”

Section: Limitationsmentioning

confidence: 99%

CDST: A Toolkit for Testing Cockpit Display Systems

Sartaj

Iqbal

Khan

2020

2020 IEEE 13th International Conference on Software Testing, Validation and Verification (ICST)

View full text Add to dashboard Cite

Avionics are highly critical systems that require extensive testing governed by international safety standards. Cockpit Display Systems (CDS) are an essential component of modern aircraft cockpits and display information from the user application using various widgets. A significant step in the testing of avionics is to evaluate whether these CDS are displaying the correct information. A common industrial practice is to manually test the information on these CDS by taking the aircraft into different scenarios during the simulation. Given the large number of scenarios to test, manual testing of such behavior is a laborious activity. In this paper, we present a CDST toolkit that automates the testing of Cockpit Display Systems (CDS). We discuss the workflow and architecture of the tool and also demonstrates the tool on an industrial case study. The results show that the tool is able to generate, execute, and evaluate the test cases and identify 3 bugs in the case study.

show abstract

Confidence Prediction for Lexicon-Free OCR

Cited by 18 publications

References 13 publications

Learning and the Unknown: Surveying Steps toward Open World Recognition

Learning and the Unknown: Surveying Steps toward Open World Recognition

Adversarial Attacks on Binary Image Recognition Systems

CDST: A Toolkit for Testing Cockpit Display Systems

Contact Info

Product

Resources

About