Impact of Covid-19 on people and Work from Home

This pandemic has affected family life around the world. As a result of lockdown individuals are already experiencing significant revenue and job losses. The ability to ‘Work from Home’ (WFH) can help damp down the impact of the situation, undoubtedly. Overall, the effects of WFH arrangements rely a lot on the job status of parents and presence of dependents (children & elderly), and this current situation is likely to intensify these differences. It does not necessarily mean, however, that the effect of the crisis should inevitably be gender neutral. Working women have been particularly affected. It seems to be very fascinating to work from home, while sitting on a comfortable couch, casually dressed, even sometimes in sleep suits, without stepping out in scorching heat and wasting time in traveling, but this may not be a preferred situation for everybody, especially women. The most significant impact on working women during the crisis will be trying to balance household demands, childcare needs and work demands. The group most likely to be hardest hit then would be lower income families with young children, and single mothers in particular. Generally women are in charge of planning, organizing and recalling of everything that needs to be remembered. The mental stress and load that comes with such work has risen exponentially in present circumstances. Even though many countries are relied on lockdown to control widespread of COVID-19 pandemic, the mental problems such as depression, anxiety, insomnia, suicidal thoughts and other psychological trauma are most common in case of normal individual and extensive in case of people who are psychologically ill. Females are more prone to psychological distress. The main concern is to manage and provide opportunities for regulation of stress caused due to anxiety and lack of peer contact. Another main threat is an increased risk of parents to develop mental illness, women may also suffer from domestic violence and consequently it results child maltreatment. The current scenario may be particularly challenging especially for children and adolescents with special needs or disadvantages, such as disabilities, also if someone has prior trauma experiences, undiagnosed mental health problems, background of migration and low socioeconomic status.

Optical Character Recognition for Printed Urdu Nastaliq Font

Optical Character Recognition (OCR) is one of the most investigated pattern classification problems that has received remarkable research attention for more than half a century. From the simplest systems recognizing isolated digits to end-to-end recognition systems, applications of OCRs vary from postal mail sorting to reading systems in scene images facilitating autonomous navigation or assisting the visually impaired. Despite tremendous research endeavors and availability of commercial recognition engines for many scripts, recognition of cursive scripts still remains an open and challenging research problem mainly due to the complexity of script, segmentation issues and large number of classes to recognize. Among these, Urdu makes the subject of our study. More specifically, this study investigates the recognition of printed Urdu text in Nastaliq style, the most widely employed script for Urdu text that is more complex than the Naskh style of Arabic. This work presents a holistic (segmentation-free) technique that exploits ligatures (partial words) as units of recognition. Urdu has a total of more than 26,000 unique ligatures, many of the ligatures, however, share the same main body (primary ligature) and differ only in the number and position of dots and diacritics (secondary ligatures). We exploit this idea to separately recognize the primary and secondary ligatures and later re-associate the two to recognize the complete ligature. Recognition is carried out using two techniques; the first of these is based on hand-crafted statistical features using hidden Markov models (HMMs). Features extracted using sliding windows are used to train a separate model for each ligature class. Feature sequences of the query ligature are fed to all the models and recognition is carried out through the model that reports the maximum probability. The second technique employs Convolutional Neural Networks (CNNs) to automatically extract useful feature representations from the classes and recognize the ligatures. We investigated the performance of a number of pre-trained networks using transfer learning techniques and trained our own set of networks from scratch as well. Experimental study of the system is carried out on two benchmark datasets of Urdu text, the ‘Urdu Printed Text Images’ (UPTI) database and the ‘Center of Language Engineering’ (CLE) database. A number of experimental scenarios are considered for system evaluation and the realized recognition rates are compared with state-of-the-art recognition systems for printed Urdu text. An interesting aspect of experimental study is the combination of unique ligatures in the two datasets to generate a large set of around 2800 unique primary and secondary ligatures covering a major proportion of the Urdu corpus. The system reports high classification rates (88.10% and 94.78% on CLE and UPTI query ligatures respectively) demonstrating the effectiveness of the proposed recognition techniques which can be adapted for other cursive scripts as well. The findings of this study are expected to be useful for the document recognition community in general and researchers targeting cursive scripts in particular.