Beijing Engineering Center for Advanced Sensors in Agriculture

Beijing, China

Beijing Engineering Center for Advanced Sensors in Agriculture

Beijing, China
Time filter
Source Type

Zeng L.,Applied Technology Internet | Zeng L.,Agricultural University of Hebei | Zeng L.,China Agricultural University | Zeng L.,Key Laboratory of Agricultural Information Acquisition Technology | And 16 more authors.
Sensor Letters | Year: 2014

The stability of long-term monitoring equipment is important especially in aquaculture. A method and a sensor to determine the turbidity of aquaculture water are presented. The optical system consists of an IR light-emitting diode, a collimating lens, and two side-by-side photodiodes. The design takes advantage of dual detectors intensity ratio. According to Beer-Lambert law, a proportional model, turbidity with the logarithm of two detectors intensity ratio (LNT), was obtained. Laboratory testing with Formazin standard showed that LNT had a good repeatability and accuracy. Better stability was verified while instability of excitation light source. And the turbidity value was rarely affected by fouling of detector's window. A comparative study with a commercial sensor was done and it shows that similar trends and smaller standard error. Copyright © 2014 American Scientific Publishers.

Duan Y.,China Agricultural University | Duan Y.,Applied Technology Internet | Duan Y.,Key Laboratory of Agricultural Information Acquisition Technology | Duan Y.,Beijing Engineering Center for Advanced Sensors in Agriculture | And 20 more authors.
Sensor Letters | Year: 2014

Aquatic animal behavior information is vital to the effective management of commercial aquaculture facilities, because most physiological and environmental changes are capable of inducing variations in animal behavior. However, behavioral variations of aquatic animal populations are difficult to measure quantitatively because the inspected subjects are sensitive, easily stressed and free to move in the complex dynamic environment. To observe and quantify such behavioral information, many different observing methods and systems have been developed to analyze aquatic animal behavior in aquaculture. This review presents the significance and recent developments of animal behavior research in the aquaculture, the basic concepts and technologies associated with animal behavior research are discussed in detail. Specially, as a vital component of automated animal behavior monitoring and analyzing, the computer vision technology and its application in aquatic animal behavior research is discussed in detail. Copyright © 2014 American Scientific Publishers.

Zhang H.,China Agricultural University | Zhang H.,Tianjin Agricultural University | Zhang H.,Applied Technology Internet | Zhang H.,Key Laboratory of Agricultural Information Acquisition Technology | And 5 more authors.
Computers and Electronics in Agriculture | Year: 2014

Cotton is an important crop throughout the world, and its quality plays a significant role in its profitability and marketability. Foreign matter in cotton can cause damage to spinning, weaving, and dyeing and thus seriously affects the quality of cotton products. Conventional methods including inspection by human workers and instrument based approaches such as photoelectric detection and ultrasonic detection are time-consuming, labor-intensive, and sometimes inaccurate. As a non-destructive, cost-effective, rapid, and objective inspection tool, computer vision has been widely used in cotton foreign matter inspection. In this review, the basic concepts, components, and image acquisition modes of computer vision techniques are presented. The improvements in image processing and analysis of foreign matter in cotton are introduced, and several different computer vision systems that have been created to detect foreign matter are reviewed to highlight the potential for the inspection of foreign matter. Considering the progress made to solve this type of problem, we also suggest some directions for future research. © 2014 Elsevier B.V.

Yu H.,China Agricultural University | Yu H.,Applied Technology Internet | Yu H.,Key Laboratory of Agricultural Information Acquisition Technology | Yu H.,Beijing Engineering Center for Advanced Sensors in Agriculture | And 4 more authors.
Sensor Letters | Year: 2013

In recent years, Information Technologies have been used in humanized management as a new model of business management, and it is more and more respected and valued. This paper presents an intelligent message sending tools based on Short Messaging Service. An intelligent matching method is proposed to match the information and people in need. The method can classify and analyze information on corporate officers, the division of SMS classes, and eventually to develop a weighting system of SMS. The system includes two parts: GSM Modem as the hardware and the Microsoft as the software. C# language was chose as the developing language. The System has been used in one college union which has 167 people. The results shows that the method is very useful for provide the information for different people.© 2013 American Scientific Publishers. Copyright © 2013 American Scientific Publishers.

Zhang L.,China Agricultural University | Li Z.,China Agricultural University | Li Z.,Key Laboratory of Agricultural Information Acquisition Technology | Li Z.,Applied Technology Internet | And 4 more authors.
International Agricultural Engineering Journal | Year: 2016

With the development of society, voices of people on animal welfare become louder and louder. Analysis of animal behavior not only concerns the matter of animal welfare, but also the issue of economy. So, the staff of China Agricultural University Experiment Station have created a system based on J2EE in Zhuozhou Hebei province, with the aim to ensure animal welfare of pig and make the pig farm workers informed the situation constantly, reducing personal subjectivity in manual inspection and making up for the defect that the pig behavior could not be recorded continuously and accurately before. The system employs computer vision technology and wireless sensing technology to access to the information of pig, setting up the pig feeding model and analyzing the nutrition and behavior of it. The system consists of four modules of multi-modal information collection module, nutritional analysis module, video image tracking module, and animal behavior analysis module. The system eliminates the environmental matters in the barn and nutritional problems of pigs timely, so as to achieve the purpose of real-time analysis. The abnormal behavior of pigs, nutritional deficiencies and other problems can be carried on the statistics and analysis synchronously through eliminating the effects of environmental factors and adjusting the pig breeding environment timely to make sure pigs grow in a healthy environment, improving the animal welfare of pigs and improving pig performance and health status, with the aim to enhance the competitive power of China's pig industry and improve the quality of pork products.

Hu J.,China Agricultural University | Hu J.,Applied Technology Internet | Hu J.,Key Laboratory of Agricultural Information Acquisition Technology | Hu J.,Beijing Engineering Center for Advanced Sensors in Agriculture | And 15 more authors.
Computers and Electronics in Agriculture | Year: 2012

Remote diagnose of fish diseases for farmers is unrealized in China, but use of mobile phones and remote analysis based on image processing can be feasible due to the widespread use of mobile phones with camera features in rural areas. This paper presents a novel method of classifying species of fish based on color and texture features and using a multi-class support vector machine (MSVM). Fish images were acquired and sent by smartphone, and the method utilized was comprised of the following stages. Color and texture subimages of fish skin were obtained from original images. Color features, statistical texture features and wavelet-based texture features of the color and texture subimages were extracted, and six groups of feature vectors were composed. LIBSVM software was tested using leave-one-out cross validation to find the best group for classification in feature selection procedure. Two multi-class support vector machines based on a one-against-one algorithm were constructed for classification. The feature selection results showed that the Bior4.4 wavelet filter in HSV color space achieved greater accuracy than the other feature groups. The classification results indicate that only the DAGMSVM meets the requirement of time efficiency for the system. The results of this study suggest that the best classification model for fish species recognition is composed of a wavelet domain feature extractor with Bior4.4 wavelet filter in HSV color space and a one-against-one algorithm based DAGMSVM classifier. © 2012 Elsevier B.V.

Loading Beijing Engineering Center for Advanced Sensors in Agriculture collaborators
Loading Beijing Engineering Center for Advanced Sensors in Agriculture collaborators