Junlong Fang

Color Analysis of Leaf Images of Deficiencies and Excess Nitrogen Content in Soybean Leaves

Abstract-Deficiencies and excess of soybeans plant nitrogen is the key of soybean nutritional diagnosis. Applying image processing technique, this paper studied the leaf images of the six stages of soybean growth, which nitrogen fertilizer applied were 0%, 50%, 100% and 150%. Using image preprocessing, the noise of source image was removed and the areas of interest were enhanced, leaves and background were separated using minimum error threshold method. The color characters of soybean leaves were analyzed using the RGB and the HSI model. The standard of distinction between normal and excessive loss of soybean leaves was found, which proved technical basis of the diagnosis of deficiencies and excess of soybeans plant nitrogen.

Design and Implementation of a Low-Power ZigBee Wireless Temperature Humidity Sensor Network

The key technology of greenhouse facilities is the monitoring of environmental parameters. Now, monitoring system of greenhouse is based on wire transmission. It is complicated to route wire and difficult to maintain. Also its reliability and anti-interference performance will degrade because of heat, light and acid. This paper leads a low-power and short range ZigBee technique into greenhouse monitoring system. In order to compose intelligent network sensor system, the paper analyses the composition of network nodes power consumption, proposes low-power design method both in hardware and software. The paper selects CC2430 module which composed of transceiver and microprocessor, and use SHT15 temperature humidity sensor. After hardware and software debugging, this wireless network can acquire and transmit data of greenhouse temperature and humidity accurately and rapidly, and this system resistant to stable work, tight structure, large loads and low power consumption.

Application of Genetic Algorithm (GA) Trained Artificial Neural Network to Identify Tomatoes with Physiological Diseases

We synthetically applied computer vision, genetic algorithm and artificial neural network technology to automatically identify the tomatoes that had physiological diseases. Firstly, the tomatoes’ images were captured through a computer vision system. Then to identify cavernous tomatoes, we analyzed the roundness and detected deformed tomatoes by applying the variation of fruit’s diameter. Secondly, we used a Genetic Algorithm (GA) trained artificial neural network. Experiments show that the above methods can accurately identify tomatoes’ shapes and meet requests of classification; the accuracy rate for the identification for tomatoes with physiological diseases was up to 100%. tomato with physiological disease; computer vision; artificial neural network; genetic algorithms

Research on the Spatial Variability of Soil Moisture Based on GIS

With the help of GPS and measuring instrument of soil moisture, soil moisture was measured and analyzed. As using Geo-statistics to the study of spatial variability of soil moisture and use ArcGIS 9.0, get the spatial distribution map of soil water property with Kriging interpolation. The research result showed that all soil spatial characters are normal distribution and the spatial distribution of soil water property accord with the fact. Geo-statistics Methods is the most appropriate methods in all of Mathematical Methods for Geostatistics. The spatial distribution map of soil water property what got with Kriging interpolation can make the spatial distribution of the entire plot, more accurate and reliable. Getting a veracious spatial distribution map of soil water speciality was very important and useful for adjusting precision fertilization and precision irrigation in time. It also offered the theoretical foundation of the connection studying between soil water speciality and enhancing the yield.

Color Analysis of Soybean Leaves Based on Computer Vision

In this paper we collected the leaves of soybean at the different stages and different nitrogen. this paper studied the leaf images of the six stages of soybean growth, which nitrogen fertilizer applied were 0%,50%,100% and 150%.The current state of development, application and research of the image processing technology are analyzed in this thesis. The color characters of soybean leaves were analyzed using the RGB and the HSI model. Get mean value of RGB and HIS, it will provide a technical basis for the diagnosis of deficiencies and excess of nitrogen for soybean leaves.

A Wireless Real-Time Monitoring Node of the Physiological Signals for Unrestrained Dairy Cattle Using Wireless Sensor Network

A newly developed smart sensor node that can monitor physiological signals for unrestrained dairy cattle is designed through modular design and its advantages are compact structure and small volume. This sensor node is based on a MSP430F133 micro-controller; the digital sensor includes temperature sensor (DS18B20-America) and vibration-displacement sensor (DN series China); transmission of the digital data uses the nRF903. The results show that this node can collect physiological signals for unrestrained dairy cattle and then send it to upper network node. This research can provide better hardware platform for further researching the communication protocols of wireless sensor networks.

Research on Modulation Technique of Novel AC/DC/AC Voltage Source Converter

This paper proposes a novel AC/DC/AC voltage source converter without or with a quite small dc-link capacitor. Because an advanced modulation wave reconstruction-SPWM (AMWR-SPWM) technique is adopted which can greatly eliminate the harmonics of output voltage, the DC filter capacitor is greatly decreased or even removed. In addition, the size of the input ac filter and the output ac filter is reduced. The principle of operation and harmonics elimination of the novel converter topology are elaborated. A through analysis on its performance is presented. This converter has many advantages such as simpler structure, higher reliability, more effective harmonics elimination. The performance of this converter using AMWR-SPWM technique is compared with traditional inverter by simulation, and the results show that the theoretical analysis is correct.

Research on Regional Spatial Variability of Soil Moisture Based on GIS

As one of soil dynamics properties, soil moisture content is an important factor of soil fertility which counts for much to crop growth situation and scientific irrigation management. A design plan of regional spatial variation of soil moisture measurement was introduced. Its main job includes the use of differential GPS technology for each sampling points in farmland, collecting data of high-precision geo-spatial information and soil moisture in farmland resorting on measure instruments of soil moisture, communicating the data between measuring instrument and portable data analysis devices or computer with cable or wireless network based on ZigBee technology, analyzing data of experimental farmland of the topography and terrain, processing and interpolating data of soil moisture content.

EXPERIMENTAL STUDY FOR AUTOMATIC COLONY COUNTING SYSTEM BASED ONIMAGE PROCESSING

Colony counting in many colony experiments is detected by manual method at present, therefore it is difficult for man to execute the method quickly and accurately .A new automatic colony counting system was developed. Making use of image-processing technology, a study was made on the feasibility of distinguishing objectively white bacterial colonies from clear plates according to the RGB color theory. An optimal chromatic value was obtained based upon a lot of experiments on the distribution of the chromatic value. It has been proved that the method greatly improves the accuracy and efficiency of the colony counting and the counting result is not affected by using inoculation, shape or size of the colony. It is revealed that automatic detection of colony quantity using image-processing technology could be an effective way.

Near Infrared Spectrum Detection of Soybean Fatty Acids Based on GA and Neural Network

This paper represented a way to build mathematical model on genetic multilevel forward neural network. Building the relationship between chemistry measurement values and near infrared spectrum datum. The near infrared spectrum data was input in this network, five kinds of content of fatty acids, which measured by chemistry method, were output. Training the weight of multilevel forward neural network by genetic algorithms, building the soybean fatty acids neural network detection model, and exploring the network model which can realize near infrared spectrum detection exactly and efficiently. The authors designed a multilevel forward neural network trained by genetic algorithms. Test showed that relative coefficient in five fatty acids of soybean can be round about 0.9, and can satisfy init detection of soybean breeding.