Research

Wireless Sensor Network for Agricultural Productions

    

Soil moisture monitoring:

     

 

  

    Wireless sensor network for cattle monitoring

      Wireless image sensor network for pecan weevil monitoring 

(images coming soon)

 
     
Machine Vision for Food Quality Detection

Chicken nugget texture analysis     

                 

Ground berry grading

  

••

     

Hyperspectral Imaging combines conventional digital imagery with spectroscopy and provides not only spatial information, but also spectral information at each pixel in the image, which can be  used to analyze responses at various wavelengths in the visible and infrared wavebands to ascertain minor and/or subtle physical and chemical features in an object.

 

Pork quality evaluation

Early detection of Apple bruise
     

Electronic Sensory Techniques

Electronic noses:
“An electronic nose is an instrument, which comprises of an array of electronic chemical sensors with partial specificity and an appropriate pattern-recognition system, capable of recognizing simple or complex odors.”

 

Maple syrup flavor evaluation

Mango maturity prediction
     
Robotic navigation

     

Optical weed sensor

  

The weed sensor was designed based on five feature wavelengths selected based on plant spectral characteristics. Four color indices compensated for the dark-current effect of phototransistors were used to develop weed classification models.  

     

A distributed, real-time, embedded weed -detection and spray-control system that integrates two weed sensors; three microcontrollers, each containing four types of peripheral modules – analog, digital, serial communication, and pulse-width modulation; a global-positioning system (GPS) receiver; a spray unit; a radar ground sensor, and an optional PC computer. A Controller Area Network (CAN) was used for simple and effective communication among the microcontrollers. The complete system was tested in wheat fields. In general, herbicide-spray accuracy achieved 80%.

A real-time, embedded weed detection and spray control system

 

    

[click on picture]            [click on picture]

 
     

Detecting Vitreousness of Durum Wheat

 

            GrainCheck310                    Cervitec

 

Manufacturer website: http://www.foss.dk/c/p/

 Two designs, one handing multi-kernels and the other handling single kernels, were tested and compared. Several calibration models were developed using neural network to classify vitreous durum kernels and several types of damaged kernels. This system provided a highly useful tool for objective grain grading, on-line measurement, and end-use property assessment of single kernels or bulk grain samples.

 

 [Home]      [Research]     [Teaching]     [Publications]     [Links]     [Others]   

[OSU]   [DASNR]    [CEAT]    [BAE]   [My Official Website]