基于FPGA的人体静脉血管成像仪的设计与实现

发布时间：2018-01-19 20:22

本文关键词： 人体静脉血管近红外成像血管增强血管提取　出处：《安徽大学》2017年硕士论文　论文类型：学位论文

【摘要】：随着科技的不断进步与发展,不但人们的生活有了巨大的变革,并且科技的快速进步对医学中病理的辅助与分析起到较大的促进作用。人体静脉血管成像技术作为一种比较前沿的近红外数字成像技术,可以有效的解决压脉注射、静脉曲张、静脉术后修复与观察等需要清晰静脉纹路信息的医学问题。本文是利用静脉血管与其周围对近红外光的吸收和反射能力的不同来实现人体静脉血管的成像。这种方便快捷无创伤的非接触式静脉成像仪可以清晰地显像不同脂肪层厚度、肤色的人体静脉血管纹路信息。人体静脉血管成像仪由FPGA控制单元、MT9V034图像采集、CY7C68013图像数据传输、映射到微处理器上的图像处理算法、双端口图像处理缓存SDRAM五部分组成。在850nm近红外光的作用下,FPGA获取MT9V034上的图像数据,通过CY7C68013将处理之后的人体静脉血管图像传输到上位机,控制和数据总线完全独立的两片SDRAM用于处理并缓存静脉图像数据。人体静脉血管成像仪上位机软件实现了人体静脉血管图像数据的接收、静脉血管图像的增强和静脉血管纹路的提取。850nm近红外光作为人体静脉成像仪的光源。根据不同人体光照强度的需要,采用PWM调光快速有效的调节850nm发光LED的亮度以实现高质量的静脉图像。在静脉采集中,光照强度可调节是十分重要的,不仅可以获得最佳的光照强度及均匀的光强分布,还可以获取较高质量的静脉图像。在这两点兼具的同时还可以减少不必要的电光线。人体静脉血管图像的捕获需要由对近红外光谱敏感的图像传感器完成,在本文中图像传感器芯片采用镁光的MT9V034,该芯片具有感光效果好;光谱效应对近红外光敏感即可以快速成像850nm的单光谱特性;成像噪声小可以很好的一直;高灵敏度;低噪声;分辨率的优点。分辨率越高,图像的细节信息越丰富、图像越清晰,即图像质量越高。USB(Universal Serial Bus)通用串行总线是一种通用数据传输规范,具有数据传输速率快(高达480Mb/s)、热插拔、操作便捷等优点。该接口是采用双差分信号线进行信号的传输,在传输过程中信号的准确性和工作效率较高。图像数据本身数据量大,在静脉采集中需要连续获取高质量的静脉血管图像。静脉血管图像是红外图像的一种,在采集过程中,图像存在的目标与背景对比度较差、图像边缘信息模糊、含有噪声较大等特点,静脉图像由于本身及成像条件的限制,图像的质量不高(包括含有较大噪声、图像对比度不高、光照不均匀、受肤色脂肪层等的影响静脉显示效果不明显)。便捷无损地获取人体静脉图像,通过图像处理技术的手段提高图像质量来辅助医疗信息的获取。本文根据人体静脉成像的原理和各部分器件选取的特殊性,设计了一种在近红外光的作用下,利用FPGA读取MT9V034图像传感器捕获由人体皮肤作用后反射的近红外光形成的人体图像数据,并在FPGA内部进行图像数据的增强、去噪处理以及将处理之后的图像通过告诉USB芯片CY7C68013传送到静脉提取上位机中。在上位机中采用的数字图像处理技术有基于DCT域的静脉图像去噪、基于YCbCr颜色空间的静脉图像肤色检测、基于直方图均衡化和Retinex的静脉图像增强、基于最大邻域内差的静脉血管提取和基于灰度共生矩阵的静脉图像质量评估的图像处理算法。经过处理之后获取清晰的人体静脉血管纹路信息。在本文中还涉及电源电路、850nm光源与调光电路、MT9V034图像采集电路、CY7C68013传输电路、FPGA电路等电路的设计以及软件代码的开发工作。最终设计了一套软硬件配合的人体静脉血管成像仪。成像仪具有两种方式处理人体静脉血管图像的方式,一种是实时接收静脉成像仪的图像数据,另一种是获取保存的静脉图片。前一种方式为实时自动处理静脉图像并提取静脉血管,后一种是手动处理静脉图像并提取静脉血管。
[Abstract]:With the continuous development of science and technology, not only the people's life has been a huge change, and assisted with analysis of the rapid progress of science and technology of medical pathology plays a larger role in promoting human vein imaging technology. As a kind of near infrared digital imaging technology is advanced, and can effectively solve the problems of pressure pulse injection, intravenous varicose vein surgery, medical problems after repair and observation need clear vein information. This article is the use of vein and its surrounding absorption and reflection ability of near-infrared light to achieve the different imaging of human vein. This convenient non-contact vein imaging instrument without trauma can clearly display the different fat thickness. Human vein lines of color. Human vein imaging by the FPGA control unit, MT9V034 image, CY7C68013 image data transmission, mapping to The image processing algorithm on the microprocessor, double port image processing SDRAM cache is composed of five parts. In the 850nm near infrared light under the action of FPGA to obtain the image data on the MT9V034, through CY7C68013 after the processing of human vein image transmission to the host computer, two pieces of SDRAM control and data bus is completely independent of the buffer for vein image data processing and human vein imaging instrument. The PC software can receive human vein image data source, extract.850nm vein image enhancement and venous lines as human vein near infrared imager. According to the different light intensity of human needs, the brightness of the PWM dimming fast and effective regulation of 850nm light emitting LED in order to achieve high quality image. In the vein vein acquisition, adjustable light intensity is very important, not only can obtain the best light intensity The light intensity distribution and uniform, vein image can also obtain high quality. At the same time it can also reduce both electro-optic line unnecessary. Capture of human vein images need to be completed by the image sensor is sensitive to the near infrared spectrum, the image sensor chip using flash MT9V034, the chip has the photosensitive effect is good; the spectral effect of near infrared light sensitive to 850nm fast imaging of single spectral imaging characteristics; low noise can have been very good; high sensitivity; low noise; resolution advantages. The higher the resolution, the image details of the more abundant information, the image is more clear, the image quality is higher (.USB Universal Serial Bus) is a general and universal serial bus data transmission standard, with fast data transfer rate (up to 480Mb/s), hot swappable, convenient operation and other advantages. The interface is using double differential signal line The signal transmission, the signal transmission in the process of accuracy and high working efficiency. The image data itself a large amount of data, vein image acquisition in vein need continuous access to high quality. The vein image is a kind of infrared image, in the image acquisition process, the contrast of target and background is the image edge fuzzy information, containing noise characteristics, because the vein image itself and the imaging conditions, the image quality is not high (including with strong noise, the image contrast is not high, uneven illumination, the color effect of intravenous fat layer of the display effect is not obvious). The convenient non-destructive testing of human vein image acquisition. To assist the medical information by means of image processing technology to improve the image quality. According to the special principle of human vein imaging and the selection of devices, a near red design Light under the action of using FPGA to read MT9V034 image sensor to capture human body image data formed by near infrared reflectance of human skin, and enhance the image data in the FPGA, denoising processing and image processing after by telling USB chip CY7C68013 transmitted to the host computer. Vein extraction of digital image processing technology the PC in the vein image denoising based on DCT domain, the vein image color detection based on YCbCr color space, histogram equalization and vein image enhancement based on Retinex, based on the veins in the neighborhood of difference and maximum extraction based on image processing algorithm of gray level co-occurrence matrix vein image quality assessment after treatment. To obtain human vein lines clear. This paper also relates to the power supply circuit, 850nm light source and light adjusting circuit, MT9V034 image collecting circuit CY7C68013, transmission circuit, FPGA circuit design and software development of the code. The final design a set of hardware and software with the human vein imaging. Imaging instrument has two modes of human vein image processing, a real-time image data receiving vein into the instrument, the other is a vein image acquisition preservation. The former approach for real-time automatic processing of image and vein vein extraction, after a manual processing and image vein vein extraction.

【学位授予单位】：安徽大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TH77;TP391.41

【相似文献】