CAPTCHA Recognition Using Deep Learning with Attached Binary Images

Thobhani, Alaa; Gao, Mingsheng; Hawbani, Ammar; Ali, Safwan; Abdussalam, Amr

doi:10.3390/electronics9091522

Cited by 19 publications

(12 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, CAPTCHA image segmentation has been implemented using the vertical projection [29][30][31] approach based on the work of Zhang et al [32]. In their work, they enhanced the vertical projection to treat the characters with a combination of size features of the characters and their locations with a vertical projection histogram [3]. The connected component algorithm has been used to segment Yahoo and Google CAPTCHA schemes [5].…”

Section: Discussionmentioning

confidence: 99%

“…The connected component algorithm has been used to segment Yahoo and Google CAPTCHA schemes [5]. However, according to Thobhani et al [3], the vertical projection and connected component algorithms require massive preprocessing procedures, which are computationally expensive and consume a lot of time.…”

Section: Discussionmentioning

confidence: 99%

“…In this kind of network architecture, the CNN serves as the feature extractor to obtain meaningful information from the CAPTCHA image such as feature vectors, and a variant of RNN, such as the Long-Short Term Memory (LSTM) network, which is used to transform the feature vectors into a text sequence. Even though this kind of model has a high recognition rate, according to Thobhani et al [3], the architecture of the CNN-RNN model is relatively complicated and could also result in increased memory and storage size.…”

Section: Discussionmentioning

confidence: 99%

“…In the work of Thobhani et al [3], they proposed an attached binary image algorithm. In their ABI algorithm, they made a specific number of copies of the input CAPTCHA image, which is equal to the number of characters in the input CAPTCHA image.…”

Section: Discussionmentioning

confidence: 99%

“…[2]. The majority of text-based CAPTCHAs consist of English uppercase letters (A to Z), English lowercase letters (a to z), and numerals (0 to 9) [3]. These text-based CAPTCHAs are distorted text images which can be misrecognized by computers or robots but can highly be recognized by humans.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

An Efficient and Accurate Depth-Wise Separable Convolutional Neural Network for Cybersecurity Vulnerability Assessment Based on CAPTCHA Breaking

Dankwa

Yang

2021

Electronics

View full text Add to dashboard Cite

Cybersecurity practitioners generate a Completely Automated Public Turing test to tell Computers and Humans Apart (CAPTCHAs) as a form of security mechanism in website applications, in order to differentiate between human end-users and machine bots. They tend to use standard security to implement CAPTCHAs in order to prevent hackers from writing malicious automated programs to make false website registrations and to restrict them from stealing end-users’ private information. Among the categories of CAPTCHAs, the text-based CAPTCHA is the most widely used. However, with the evolution of deep learning, it has been so dramatic that tasks previously thought not easily addressable by computers and used as CAPTCHA to prevent spam are now possible to break. The workflow of CAPTCHA breaking is a combination of efforts, approaches, and the development of the computation-efficient Convolutional Neural Network (CNN) model that attempts to increase accuracy. In this study, in contrast to breaking the whole CAPTCHA images simultaneously, this study split four-character CAPTCHA images for the individual characters with a 2-pixel margin around the edges of a new training dataset, and then proposed an efficient and accurate Depth-wise Separable Convolutional Neural Network for breaking text-based CAPTCHAs. Most importantly, to the best of our knowledge, this is the first CAPTCHA breaking study to use the Depth-wise Separable Convolution layer to build an efficient CNN model to break text-based CAPTCHAs. We have evaluated and compared the performance of our proposed model to that of fine-tuning other popular CNN image recognition architectures on the generated CAPTCHA image dataset. In real-time, our proposed model used less time to break the text-based CAPTCHAs with an accuracy of more than 99% on the testing dataset. We observed that our proposed CNN model has efficiently improved the CAPTCHA breaking accuracy and streamlined the structure of the CAPTCHA breaking network as compared to other CAPTCHA breaking techniques.

show abstract