Automatic Visual Features for Writer Identification: A Deep Learning Approach

Abstract

Identification of a person from his writing is one of the challenging problems; however, it is not new. No one can repudiate its applications in a number of domains, such as forensic analysis, historical documents, and ancient manuscripts. Deep learning-based approaches have proved as the best feature extractors from massive amounts of heterogeneous data and provide promising and surprising predictions of patterns as compared with traditional approaches. We apply a deep transfer convolutional neural network (CNN) to identify a writer using handwriting text line images in English and Arabic languages. We evaluate different freeze layers of CNN (Conv3, Conv4, Conv5, Fc6, Fc7, and fusion of Fc6 and Fc7) affecting the identification rate of the writer. In this paper, transfer learning is applied as a pioneer study using ImageNet (base data-set) and QUWI data-set (target data-set). To decrease the chance of over-fitting, data augmentation techniques are applied like contours, negatives, and sharpness using textline images of target data-set. The sliding window approach is used to make patches as an input unit to the CNN model. The AlexNet architecture is employed to extract discriminating visual features from multiple representations of image patches generated by enhanced pre-processing techniques. The extracted features from patches are then fed to a support vector machine classifier. We realized the highest accuracy using freeze Conv5 layer up to 92.78% on English, 92.20% on Arabic, and 88.11% on the combination of Arabic and English, respectively. INDEX TERMS Writer identification, visual features, AlexNet, multilingual, support vector machine.

RETRACTED ARTICLE: Refining Parkinson’s neurological disorder identification through deep transfer learning

“…In addition to the relatively small sample size, the other important limitation of most of the studies are focused on hand crafted features due to limitation of deep learning methods on small dataset, however, automatic extraction of features could helps to increase the identification performance [13]. Deep learning methods have demonstrated tremendous success in a variety of applications in various fields [14][15][16][17][18][19], however, it is data hungry approach and requires atleast 10 times the degree of freedom that can often preclude the use of CNNs for applications where dataset can be challenging. In order to address the problem of limited training data, transfer learning could be used to tune the already grained storing knowledge on similar problem.…”

FragNet: Writer Identification Using Deep Fragment Networks

“…In [34], a denoising network is used to extract deep features on small patches. A transfer deep learning from ImageNet is used in [35] where deep features are extracted on small image patches and then fed to a SVM classifier. Keglevic et al [36] apply a triplet network to learn a similarity measure for image patches and the global feature is computed as the vector of locally aggregated image patch descriptors.…”

Deep OCR for Arabic script‐based language like Pastho

“…The deep learning technologies enable computers to take human‐like decisions and make predictions. Deep learning has been successfully applied to many applications like face recognition, speech recognition, text recognition (Naz et al, 2017), writer identification (Rehman, Naz, Razzak, & Hameed, 2019), medical disease predictions (Naseer et al, 2020; Rehman, Naz, Razzak, Akram, & Imran, 2020), machine translation, computer vision, natural language processing, etc. The CNNs deep learners have proved a tremendous success for pattern recognition applications and technologies.…”