Design And Development Of Intelligent Video Magnifier Software Based On DM8127

As the most precious sensory organ of mankind,the eye is the most important part of human access to external information.Low vision group as an important part of the visually impaired people,their quality of life is seriously affected by low vision problems,which not only bring physical and psychological harm,but also cause great distress to their families.Nowadays,low vision problem has been paid more and more attention in the field of rehabilitation.Through the appropriate electronic video magnifiers to improve the residual vision is an effective way to meet the need of daily life of learning and work.In this thesis,a set of intelligent video magnifier software is designed and developed based on the TI's high-performance multi-core video processor DM8127.The main work and achievements of this thesis are as follows:(1)This thesis illustrates the background and the significance of this subject,analyzes the research status of electronic video magnifier at home and abroad,introduces the application of embedded multimedia processor,and summarizes the main research content and structure of this thesis.(2)The DM8127 multi-core video processor's software development framework McFW(Multi-Channel FrameWork)is analyzed,as well as the most important Link working mechanism.Based on the different functions of various Link,including image capture,display,scaling,discoloration and other processing steps,the basic function of video magnifier software is realized.(3)Through the Google open source OCR(Optical Character Recognition)engine,the character information in the captured image is identified and extracted successfully.By using Freetype font engine to achieve the font outline extraction from vector font file,combined with the recognized characters,the different font size images are reconstructed.Finally,the TTS(Text to Speech)software library is used to realize the conversion from characters to speech,so as to realize the visual + auditory text information acquisition for the low vision group.(4)Based on the DM8127 video codec module in the M3-VIDEO core,the camera/video recording save function is realized by the encode Link,the file playback function is realized by the decodeLink.Based on the FTP server and curl network tool,the file network transmission function is designed to achieve the file upload and download functions.(5)Finally,the content of this thesis is summarized,and the future research direction is also prospected.