基于ZYNQ图像处理研制便携式水质检测系统  

DEVELOPMENT OF A PORTABLE WATER QUALITY DETECTION SYSTEM BASED ONZYNQ IMAGE PROCESSING

作　　者：孙冰洋 杨顺生[1] 陈鹏[1] 张大文 SUN Bingyang;YANG Shunsheng;CHEN Peng;ZHANG Dawen(School of Civil Engineering,Southwest Jiaotong University,Chengdu 610031,China)

机构地区：[1]西南交通大学土木工程学院,成都610031

出　　处：《环境工程》2023年第7期175-183,234,共10页Environmental Engineering

摘　　要：综合光电检测技术和光谱分析技术中的紫外-可见吸收光谱法,研制多参数便携式地表水水质检测系统,能够现场快速检测出磷酸盐、亚硝酸盐、化学需氧量(COD)和NH3-N水质参数。对水体中吸收特征波长在可见光部分的物质,使用摄像头采集其可见光谱,并对其可见光谱图像的灰度图进行卷积神经网络建模,吸收特征波长在紫外波段的物质,通过光电检测技术测其浓度值,将建立的卷积神经网络模型移植到ZYNQ中,结合紫外光电传感器,将被检测物质的浓度值通过LCD显示出来,以此实现水质检测仪的便携性。结果表明:所得卷积神经网络预测值为样本溶液在8种浓度值输出类型中的倾向值,准确率最高为100%,最低为40%。COD浓度值的最高误差为10%,证实该检测系统具有很好的实用价值。Combined with photoelectric detection technology and UV-Vis absorption spectroscopy in spectral analysis technology,a multi-parameter portable surface water quality detection system was developed,which can quickly detect phosphate,nitrite,and chemical oxygen demand(COD)and ammonia nitrogen on-site.For the substances in the water body that absorb the characteristic wavelengths in the visible range,a camera was used to collect the visible spectrum,and the grayscale image of the visible spectrum image was modelled by a convolutional neural network.The concentration value of substances,whose absorption characteristic wavelength is within the ultraviolet band,was measured by photoelectric detection technology.The established convolutional neural network model was transplanted into ZYNQ,and combined with an ultraviolet photoelectric sensor,the concentration value of the detected substance was displayed on the LCD to realize the portability of the water quality detector.Research indicated that:the prediction value of the convolutional neural network was obtained as the tendency value of sample solution in 8 output types of concentration value,the highest accuracy was 100%,and the lowest was 40%.The highest error of COD concentration value was 10%,proving that the detection system has good practical value.

关 键 词：紫外-可见吸收光谱法 水质检测 卷积神经网络 灰度模型 

分 类 号：O657.3[理学—分析化学] X84[理学—化学]