This Braille reader technology incorporates innovations to enhance accessibility.
Design and Functionality of the Sensor
The new sensor contains an optical fibre ring resonator that is embedded in soft PDMS material that acts like a flexible optical skin.
The sensor can measure through applied pressure and results in frequency changes in light and converts them into readable data.
The researchers used a technique called Pound-Drever-Hall (PDH) frequency locking to convert the small frequency changes into readable data.
This approach keeps the light signal stable even in dynamic environments so that readings are accurate no matter what else is happening around the sensor.
Enhancing Accuracy Through Machine Learning
Machine learning was also used to enhance the sensor's accuracy. A Multilayer Perceptron Neural Network was used to help with the recognition of specific Braille patterns.
A Long Short-Term Memory network turned sequences of Braille into readable texts or words.
When combined with neural network data processing methods, the system could quickly read Braille letters, numbers, and punctuation.
Through testing, the sensor identified eight different Braille patterns with 98.57% accuracy.
Improved Precision and Future Plans
The sensor could respond to pressure in less than 0.1 seconds and could read an entire Braille word correctly, even with slight variations in each character.
“This system is far more precise than older Braille reader technology that might miss imperfectly pressed dots,” said Rui Min from Beijing Normal University, co-author of the research paper on this sensor. “The flexible optical fibre resonator detects very small pressure differences, and PDH frequency locking ensures stability and accuracy, even with changing light or power fluctuations. Machine learning further enhances the system, allowing it to recognize Braille despite minor errors or pressure variations.”
The team plan to make the sensor for Braille reader technology more robust for real-world use by optimising it for different devices and reducing costs.
