Japanese Braille Translation Using Deep Learning - Conversion from Phonetic Characters (Kana) to Homonymic Characters (Kanji).

A blind student writes and submits reports in Braille word processor, which is difficult for teachers to read. This study's purpose is to make a translator from Braille into mixed Kana-Kanji sentences for such teachers. Because Kanji has homonyms, it is not always possible to get correct results when converting. To overcome this difficulty, we used deep learning for translation. We built a training dataset composed from 15,000 pairs of Braille codes and mixed Kana-Kanji sentences, and a validation dataset. In training, we got an accuracy of 0.906 and a good Bleu score of 0.600. In validation, we found 5 mistaken words in selecting homonymous Kanji by examining translation mistakes from 100 pairs of the verification sentences. The choice of homonymous Kanji depends on the context. For decreasing such type of errors, it is necessary to introduce of translation of paragraphs by increasing the scale of the network model in deep learning, and to expand the network structure.

Efficacy of low-vision devices in elderly population with age-related macular degeneration.

Purpose: Age-related macular degeneration (AMD) is a common cause of blindness, residual damage to macular area in spite of treatment necessitates visual rehabilitation by means of low-vision aids (LVAs). Methods: Thirty patients suffering from different stages of AMD requiring LVAs were included in this prospective study. Patients with nonprogressive, adequately treated AMD were enrolled over a 12-month period, prescribed requisite LVAs and followed-up for a minimum 1-month period. Before and after provision of LVAs, near work efficiencies were evaluated by calculating reading speed as words per minute (wpm) under both photopic and mesopic light conditions, and impact of poor vision on activities of daily living (ADL) was quantified by modified standard questionnaire based on Nhung X et al. questionnaire. Results: Of the 30 patients mean studied with mean age of 68 ± 10 years, 20 patients (66.7%) had dry AMD in better eye and 10 (33.3%) had wet AMD. Post-LVA, near visual acuity improved significantly and all cases were able to read some letters on near vision chart with an average improvement of 2.4 ± 0.96 lines. The different LVAs prescribed were high plus reading spectacles (up to 10 D) in 23.3%, hand-held magnifiers in 53.3%, base in prisms in 10%, stand held magnifiers in 6.7%, and bar and dome magnifiers in 3.3%. Conclusion: LVAs are effective in visual rehabilitation in patients with AMD. Self-reported reduction in visual dependency and improvement in vision-related quality of life post use of aids corroborated perceived benefit.

Multi-Sensor Data Fusion Solutions for Blind and Visually Impaired: Research and Commercial Navigation Applications for Indoor and Outdoor Spaces.

Several assistive technology solutions, targeting the group of Blind and Visually Impaired (BVI), have been proposed in the literature utilizing multi-sensor data fusion techniques. Furthermore, several commercial systems are currently being used in real-life scenarios by BVI individuals. However, given the rate by which new publications are made, the available review studies become quickly outdated. Moreover, there is no comparative study regarding the multi-sensor data fusion techniques between those found in the research literature and those being used in the commercial applications that many BVI individuals trust to complete their everyday activities. The objective of this study is to classify the available multi-sensor data fusion solutions found in the research literature and the commercial applications, conduct a comparative study between the most popular commercial applications (Blindsquare, Lazarillo, Ariadne GPS, Nav by ViaOpta, Seeing Assistant Move) regarding the supported features as well as compare the two most popular ones (Blindsquare and Lazarillo) with the BlindRouteVision application, developed by the authors, from the standpoint of Usability and User Experience (UX) through field testing. The literature review of sensor-fusion solutions highlights the trends of utilizing computer vision and deep learning techniques, the comparison of the commercial applications reveals their features, strengths, and weaknesses while Usability and UX demonstrate that BVI individuals are willing to sacrifice a wealth of features for more reliable navigation.

Obstacle Detection System for Navigation Assistance of Visually Impaired People Based on Deep Learning Techniques.

Visually impaired people seek social integration, yet their mobility is restricted. They need a personal navigation system that can provide privacy and increase their confidence for better life quality. In this paper, based on deep learning and neural architecture search (NAS), we propose an intelligent navigation assistance system for visually impaired people. The deep learning model has achieved significant success through well-designed architecture. Subsequently, NAS has proved to be a promising technique for automatically searching for the optimal architecture and reducing human efforts for architecture design. However, this new technique requires extensive computation, limiting its wide use. Due to its high computation requirement, NAS has been less investigated for computer vision tasks, especially object detection. Therefore, we propose a fast NAS to search for an object detection framework by considering efficiency. The NAS will be used to explore the feature pyramid network and the prediction stage for an anchor-free object detection model. The proposed NAS is based on a tailored reinforcement learning technique. The searched model was evaluated on a combination of the Coco dataset and the Indoor Object Detection and Recognition (IODR) dataset. The resulting model outperformed the original model by 2.6% in average precision (AP) with acceptable computation complexity. The achieved results proved the efficiency of the proposed NAS for custom object detection.

Braille recognition by E-skin system based on binary memristive neural network.

Braille system is widely used worldwide for communication by visually impaired people. However, there are still some visually impaired people who are unable to learn Braille system due to various factors, such as the age (too young or too old), brain damage, etc. A wearable and low-cost Braille recognition system may substantially help these people recognize Braille or assist them in Braille learning. In this work, we fabricated polydimethylsiloxane (PDMS)-based flexible pressure sensors to construct an electronic skin (E-skin) for the application of Braille recognition. The E-skin mimics human touch sensing function for collecting Braille information. Braille recognition is realized with a neural network based on memristors. We utilize a binary neural network algorithm with only two bias layers and three fully connected layers. Such neural network design remarkably reduces the calculation burden and, thus, the system cost. Experiments show that the system can achieve a recognition accuracy of up to 91.25%. This work demonstrates the possibility of realizing a wearable and low-cost Braille recognition system and a Braille learning-assistance system.

Value of Handheld Optical Illuminated Magnifiers for Sustained Silent Reading by Visually Impaired Adults.

SIGNIFICANCE: Vision rehabilitation providers tend to recommend handheld, illuminated optical magnifiers for short-duration spot reading tasks, but this study indicates that they are also a viable option to improve sustained, continuous text reading (e.g., books or magazines), especially for visually impaired adults who read slowly with only spectacle-based near correction. PURPOSE: The utility of handheld optical magnifiers for sustained silent reading tasks involving normal-sized continuous text could be a valuable indication that is not recognized by vision rehabilitation providers and patients. METHODS: Handheld, illuminated optical magnifiers were dispensed to 29 visually impaired adults who completed the sustained silent reading test by phone at baseline without the new magnifier and 1 month after using the magnifier. Reading speed in words per minute (wpm) was calculated from the time to read each page and then averaged across up to 10 pages or determined for the fastest read page (maximum). RESULTS: From baseline without the magnifier to 1 month with the magnifier, there was a significant improvement in mean reading speed by 14 wpm (95% confidence interval [CI], 2.6 to 24; P = .02) and for maximum reading speed by 18 wpm (95% CI, 5.4 to 30; P = .005) on average across participants. Participants who had slower baseline reading speeds without the magnifier demonstrated significantly greater improvements in mean and maximum reading speeds on average with the magnifier (95% CI, 8 to 32 [ P = .003]; 95% CI, 4 to 36 [ P = .02]). A significantly greater number of pages were read with the new magnifier than without it (Wilcoxon z = -2.5; P = .01). A significantly greater number of pages were read with the magnifier by participants who read fewer pages at baseline (95% CI, 0.57 to 5.6; P = .02) or had greater improvements in mean reading speed (95% CI, 0.57 to 5.6; P = .007). CONCLUSIONS: Many visually impaired adults read more quickly and/or read a greater number of pages after using a new magnifier for a month than compared to without it. The largest gains occurred among those with more difficulty at baseline, indicating the potential to improve reading rates with magnifiers for those with greater deficits.

IoT Enabled Intelligent Stick for Visually Impaired People for Obstacle Recognition.

This paper presents the design, development, and testing of an IoT-enabled smart stick for visually impaired people to navigate the outside environment with the ability to detect and warn about obstacles. The proposed design employs ultrasonic sensors for obstacle detection, a water sensor for sensing the puddles and wet surfaces in the user's path, and a high-definition video camera integrated with object recognition. Furthermore, the user is signaled about various hindrances and objects using voice feedback through earphones after accurately detecting and identifying objects. The proposed smart stick has two modes; one uses ultrasonic sensors for detection and feedback through vibration motors to inform about the direction of the obstacle, and the second mode is the detection and recognition of obstacles and providing voice feedback. The proposed system allows for switching between the two modes depending on the environment and personal preference. Moreover, the latitude/longitude values of the user are captured and uploaded to the IoT platform for effective tracking via global positioning system (GPS)/global system for mobile communication (GSM) modules, which enable the live location of the user/stick to be monitored on the IoT dashboard. A panic button is also provided for emergency assistance by generating a request signal in the form of an SMS containing a Google maps link generated with latitude and longitude coordinates and sent through an IoT-enabled environment. The smart stick has been designed to be lightweight, waterproof, size adjustable, and has long battery life. The overall design ensures energy efficiency, portability, stability, ease of access, and robust features.

An IoT Machine Learning-Based Mobile Sensors Unit for Visually Impaired People.

Visually impaired people face many challenges that limit their ability to perform daily tasks and interact with the surrounding world. Navigating around places is one of the biggest challenges that face visually impaired people, especially those with complete loss of vision. As the Internet of Things (IoT) concept starts to play a major role in smart cities applications, visually impaired people can be one of the benefitted clients. In this paper, we propose a smart IoT-based mobile sensors unit that can be attached to an off-the-shelf cane, hereafter a smart cane, to facilitate independent movement for visually impaired people. The proposed mobile sensors unit consists of a six-axis accelerometer/gyro, ultrasonic sensors, GPS sensor, cameras, a digital motion processor and a single credit-card-sized single-board microcomputer. The unit is used to collect information about the cane user and the surrounding obstacles while on the move. An embedded machine learning algorithm is developed and stored in the microcomputer memory to identify the detected obstacles and alarm the user about their nature. In addition, in case of emergencies such as a cane fall, the unit alerts the cane user and their guardian. Moreover, a mobile application is developed to be used by the guardian to track the cane user via Google Maps using a mobile handset to ensure safety. To validate the system, a prototype was developed and tested.

Convolutional Neural Network Based Real Time Arabic Speech Recognition to Arabic Braille for Hearing and Visually Impaired.

Natural Language Processing (NLP) is a group of theoretically inspired computer structures for analyzing and modeling clearly going on texts at one or extra degrees of linguistic evaluation to acquire human-like language processing for quite a few activities and applications. Hearing and visually impaired people are unable to see entirely or have very low vision, as well as being unable to hear completely or having a hard time hearing. It is difficult to get information since both hearing and vision, which are crucial organs for receiving information, are harmed. Hearing and visually impaired people are considered to have a substantial information deficit, as opposed to people who just have one handicap, such as blindness or deafness. Visually and hearing-impaired people who are unable to communicate with the outside world may experience emotional loneliness, which can lead to stress and, in extreme cases, serious mental illness. As a result, overcoming information handicap is a critical issue for visually and hearing-impaired people who want to live active, independent lives in society. The major objective of this study is to recognize Arabic speech in real time and convert it to Arabic text using Convolutional Neural Network-based algorithms before saving it to an SD card. The Arabic text is then translated into Arabic Braille characters, which are then used to control the Braille pattern via a Braille display with a solenoid drive. The Braille lettering triggered on the finger was deciphered by visually and hearing challenged participants who were proficient in Braille reading. The CNN, in combination with the ReLU model learning parameters, is fine-tuned for optimization, resulting in a model training accuracy of 90%. The tuned parameters model's testing results show that adding the ReLU activation function to the CNN model improves recognition accuracy by 84 % when speaking Arabic digits.

Sensor-Based Prototype of a Smart Assistant for Visually Impaired People-Preliminary Results.

People with visual impairment are the second largest affected category with limited access to assistive products. A complete, portable, and affordable smart assistant for helping visually impaired people to navigate indoors, outdoors, and interact with the environment is presented in this paper. The prototype of the smart assistant consists of a smart cane and a central unit; communication between user and the assistant is carried out through voice messages, making the system suitable for any user, regardless of their IT skills. The assistant is equipped with GPS, electronic compass, Wi-Fi, ultrasonic sensors, an optical sensor, and an RFID reader, to help the user navigate safely. Navigation functionalities work offline, which is especially important in areas where Internet coverage is weak or missing altogether. Physical condition monitoring, medication, shopping, and weather information, facilitate the interaction between the user and the environment, supporting daily activities. The proposed system uses different components for navigation, provides independent navigation systems for indoors and outdoors, both day and night, regardless of weather conditions. Preliminary tests provide encouraging results, indicating that the prototype has the potential to help visually impaired people to achieve a high level of independence in daily activities.

Refreshable Braille Display With Adjustable Cell Size for Learners With Different Tactile Sensitivity.

Braille is one of the most popular mediums of education for the blind. However, learning braille requires trainers and a lot of practice. Additionally, different individuals have different levels of tactile sensitivity at their fingertips. The tactile components get often overlooked in most braille learning devices and related studies. Our solution is a single cell refreshable braille display with six custom-made electromechanical flapper actuators. It incorporates speech functionalities to facilitate self-learning and independent operation. The cell size can be adjusted according to the learner's preference by moving the actuators. The device can provide standard braille cell dimensions and elevation as well. It is designed to help learners with different tactile perceptions improve themselves through practice and adapt to standard size braille. The operational conditions and force analysis of the braille dots were performed. Two tests were also performed with two different cell sizes to evaluate the device with several blind students. The device is very affordable and easy to maintain. It can also be used to teach braille to the sighted.

Enablers and barriers encountered by working-age and older adults with vision impairment who pursue braille training.

PURPOSE: We explored the experiences of working-age and older adults with acquired vision impairment who pursued braille rehabilitation training, and the facilitators and barriers they encountered throughout this process. METHODS: Semi-structured interviews of up to 90 min in length were conducted with 14 participants from across Canada who learned braille between the ages of 33 and 67 (Mdn = 46). Transcripts were analyzed by two researchers using interpretive phenomenological analysis. RESULTS: A variety of personal, social and institutional factors characterize the adult braille learning experience. Among these, participants highlight the role of prior identity and experience, the impact of access to resources and the cost of materials and devices needed to maintain braille skills. Findings also emphasize invisible barriers, including the role of societal perceptions towards braille, the level of support provided by family and friends, and the influence of unconscious biases towards braille and aging held by both adult learners and those around them. CONCLUSIONS: These findings provide important context to improve policies and practice in adult braille rehabilitation. As the prevalence of age-related vision impairment continues to increase, it will become imperative to understand the unique needs of working-age and older adults with acquired vision impairment who pursue braille.Implications for REHABILITATIONThis study is one of the first to explore the experiences of working-age and older adults with acquired visual impairment who pursue braille rehabilitation training.Rehabilitation professionals must take into account prior learning and reading experiences which may shape the braille learning process.Family members require greater access to resources and support during the training process.There is a significant need for public education to address societal misconceptions about braille and blindness that can lead to a reluctance to use braille.Interactions with other braille users foster more empowering definitions of braille that align with the social model understanding of disability.

An intelligent indoor guidance and navigation system for the visually impaired.

Intelligent guidance in complex environments where various procedures are required for navigation is critical to achieving mobility for the visually impaired. This study presents a newly developed software prototype with a hybrid RFID/BLE infrastructure to provide intelligent navigation and guidance to the visually impaired in complex indoor environments. The system enables the users to input their purpose via a specially designed user interface, and provides intelligent guidance through a chain of destination targets which are determined according to the inherent procedures of the environment. Path optimization is performed by adaptation of the traveling salesman problem, and real-time instantaneous instructions are provided to guide the users through the predetermined destination points. For evaluation purposes, a hospital environment is constructed as an example of a complex environment and the system is tested by visually impaired participants. The results show that the intelligent purpose selection and destination evaluation mechanism modules of the system are found to be effective by all the participants.

The association between tactile, motor and cognitive capacities and braille reading performance: a scoping review of primary evidence to advance research on braille and aging.

PURPOSE: As the prevalence of age-related visual impairment increases, a greater understanding of the physiological and cognitive capacities that are recruited during braille reading and the potential implications of age-related declines is required. METHODS: This scoping review aimed to identify and describe primary studies exploring the relationship between tactile, motor and cognitive capacities and braille reading performance, the instruments used to measure these capacities, and the extent to which age is considered within these investigations. English peer-reviewed articles exploring the relationship between these capacities and braille reading performance were included. Articles were screened by two researchers, and 91% agreement was achieved (kappa = 0.84 [0.81, 0.87], p < 0.01). RESULTS: 2405 articles were considered of which 36 met the inclusion criteria. Fifteen investigated the relationship between tactile capacities and braille reading performance, 25 explored motor capacities, and 5 considered cognitive capacities. Nineteen instruments were used to measure tactile capacity, 4 for motor dexterity, and 7 for cognitive capacity. These studies focus on younger participants and on those who learned braille early in life. CONCLUSIONS: Although this overview underscores the importance of tactile perception and bimanual reading, future research is needed to explore the unique needs of older adults who learn braille later in life.IMPLICATIONS FOR REHABILITATIONThe studies in this review underscore the importance of developing both haptic tactile perception and efficient hand reading patterns early in the braille learning process.Practitioners should consider whether specific pre-braille readiness activities can be used to address the unique needs of older adults who may experience tactile, motor or cognitive declines.Most of the studies in this review require replication before they should serve as reliable clinical guidelines; however, braille reading (like print) is a complex process that draws on multiple capacities that should be developed in unison.The studies in this review focus heavily on younger participants and on those who learned braille early in life, and highlight the need for future research on braille and aging.

Faces and words are both associated and dissociated as evidenced by visual problems in dyslexia.

Faces and words are traditionally assumed to be independently processed. Dyslexia is also traditionally thought to be a non-visual deficit. Counter to both ideas, face perception deficits in dyslexia have been reported. Others report no such deficits. We sought to resolve this discrepancy. 60 adults participated in the study (24 dyslexic, 36 typical readers). Feature-based processing and configural or global form processing of faces was measured with a face matching task. Opposite laterality effects in these tasks, dependent on left-right orientation of faces, supported that they tapped into separable visual mechanisms. Dyslexic readers tended to be poorer than typical readers at feature-based face matching while no differences were found for global form face matching. We conclude that word and face perception are associated when the latter requires the processing of visual features of a face, while processing the global form of faces apparently shares minimal-if any-resources with visual word processing. The current results indicate that visual word and face processing are both associated and dissociated-but this depends on what visual mechanisms are task-relevant. We suggest that reading deficits could stem from multiple factors, and that one such factor is a problem with feature-based processing of visual objects.

Temporal Dynamics of Brain White Matter Plasticity in Sighted Subjects during Tactile Braille Learning: A Longitudinal Diffusion Tensor Imaging Study.

The white matter (WM) architecture of the human brain changes in response to training, though fine-grained temporal characteristics of training-induced white matter plasticity remain unexplored. We investigated white matter microstructural changes using diffusion tensor imaging at five different time points in 26 sighted female adults during 8 months of training on tactile braille reading. Our results show that training-induced white matter plasticity occurs both within and beyond the trained sensory modality, as reflected by fractional anisotropy (FA) increases in somatosensory and visual cortex, respectively. The observed changes followed distinct time courses, with gradual linear FA increase along the training in the somatosensory cortex and sudden visual cortex cross-modal plasticity occurring after braille input became linguistically meaningful. WM changes observed in these areas returned to baseline after the cessation of learning in line with the supply-demand model of plasticity. These results also indicate that the temporal dynamics of microstructural plasticity in different cortical regions might be modulated by the nature of computational demands. We provide additional evidence that observed FA training-induced changes are behaviorally relevant to tactile reading. Together, these results demonstrate that WM plasticity is a highly dynamic process modulated by the introduction of novel experiences.SIGNIFICANCE STATEMENT Throughout the lifetime the human brain is shaped by various experiences. Training-induced reorganization in white matter (WM) microstructure has been reported, but we know little about its temporal dynamics. To fill this gap, we scanned sighted subjects five times during tactile braille reading training. We observed different dynamics of WM plasticity in the somatosensory and visual cortices implicated in braille reading. The former showed a continuous increase in WM tissue anisotropy along with tactile training, while microstructural changes in the latter were observed only after the participants learned to read braille words. Our results confirm the supply-demand model of brain plasticity and provide evidence that WM reorganization depends on distinct computational demands and functional roles of regions involved in the trained skill.

Millimeter-sized smart sensors reveal that a solar refuge protects tree snail Partula hyalina from extirpation.

Pacific Island land snails are highly endangered due in part to misguided biological control programs employing the alien predator Euglandina rosea. Its victims include the fabled Society Island partulid tree snail fauna, but a few members have avoided extirpation in the wild, including the distinctly white-shelled Partula hyalina. High albedo shell coloration can facilitate land snail survival in open, sunlit environments and we hypothesized that P. hyalina has a solar refuge from the predator. We developed a 2.2 × 4.8 × 2.4 mm smart solar sensor to test this hypothesis and found that extant P. hyalina populations on Tahiti are restricted to forest edge habitats, where they are routinely exposed to significantly higher solar radiation levels than those endured by the predator. Long-term survival of this species on Tahiti may require proactive conservation of its forest edge solar refugia and our study demonstrates the utility of miniaturized smart sensors in invertebrate ecology and conservation.

A Survey on Tactile Displays For Visually Impaired People.

Traditional paper documents with Braille characters and tangible graphics have obvious defects to disseminate knowledge in the information age. Information accessibility is an urgent challenge for blind individuals. Although many types of tactile displays were created for different applications, we especially focus on the tactile display for visually impaired people, which can dynamically generate tangible graphics and Braille characters, to help the blind obtain information conveniently. In this article, we present the state-of-the-art of graphic tactile displays (GTDs) and refreshable Braille displays (RBDs), then discuss their common kernel technologies about actuators and latch structures. This article summarizes the performance of typical actuators of tactile displays and analyzes the working principles of some latch structures. This article systematically summarizes latch structures of GTDs and RBDs, for the first time. Several comments in this paper will be useful to develop high-performance tactile displays for visually impaired people.