Airborne HUD's Video And Character Processing Algorithm Research And Driver Development

Airborne display system plays an important role in enhancing the flight dynamics awareness of the pilot as a man-machine interface between pilot and aircraft. This dissertation launched a further research based on the previous generation of HUD graphics and character generating and video processing system. Three aspects including video deinterlacing, character delicate display and interface&driver development are improved and optimized. The main contents are summarized as follows:(1) According to the new requirements of the system, the overall scheme of the system is designed. Upper application program analyses and PCI-E communication are implemented on the CPU board where VxWorks is transplanted. Graphics generating, character drawing, video processing and overlay display of the three are implemented on the GPU board which is based on Kintex 7.(2) Video deinterlacing algorithm is researched and optimized. A motion adaptive inter-field deinterlacing algorithm is proposed which solves both the full screen jitter phenomenon and serious sawtooth phenomenon caused by line replication algorithm in previous generation of HUD graphics and character generating and video processing system. An intra-field deinterlacing algorithm based on the duality between different resolution images is proposed, using only the current field information to complete the deinterlacing, effectively enhancing the detection and recovery performance of small angle edge direction.(3) Scheme of vector font display is designed and implemented base on TrueType. Firstly,on the PC side, FreeType font engine is transplanted to read the TrueType font file, completing operations including character glyph contour extraction, rotation and character stroke generation, then generating the system font file and programming it to the GPU side. Secondly, on the CPU side, offset address of character is generated based on the analytic results of character drawing instructions. Thirdly, on the GPU side, after loading character data from system font file, superposition of character and video are completed with the help of the character-video background fusion algorithm which is proposed in this dissertation, and then output to the display device.(4) Application interface program, device drivers, and system testing program are developed. Application interface program based on standard OpenGL is designed for the convenience of the upper application development. As to hardware accelerated graphics drawing instructions, video setting instructions and special functional area setting instructions, system instruction interface is defined for the direct call of the upper application program. PCI-E device driver is developed and both PIO communication and DMA transfer are implemented. System testing program is designed to complete the overall functional test verification of the graphics, character and video system.(5) Through the joint debugging of CPU board and CPU board, it validates that the proposed video processing algorithms, character display scheme and interface&driver program meet the functional requirements and technical specifications of the HUD system, and effectively guarantee the stable, reliable, and real-time display of the character and video information.