Exjobbsförslag från företag

Detta är ett uppsatsförslag hämtat från Nationella Exjobb-poolen. Klicka här för att komma tillbaka till samtliga exjobbsförslag.

Förslaget inkom 2006-01-11

Diploma work on image analysis as a tool for site specific weed control

There is a large demand from society to decrease the use of biocides in agriculture. The Swedish government has put up environmental goals and several of these involve the use of biocides in agriculture. Furthermore, it is well motivated to have a requirement controlled use of biocides even from a farmer perspective in an economically demanding situation. There exist studies that show that the weed often occurs in spots on cultivated land. It is well known that the soil has large influence on the occurrence of weeds and that one should expect differences in weed occurrence then the soil varies within a field. It has been shown that it is possible to decrease the use of biocides if one lets the amount of weed control the amount of herbicides applied on different parts in the field. The results from 18 different types of wheat and grain fields showed that it was possible to decrease the costs for the herbicides with 50% although costs for site specific registration of weed densities had been taken into account.

The proposed diploma work focus on determination of weeds in digital images. The main goal with this diploma work is to distinguish between background (soil), crop and weed. The amount of weed will then be used to control the amount of herbicides applied to the field at different locations. The image analysis work will involve algorithms for correction of spherical aberration (see the non quadratic frame in the image above), correction for different types of natural illuminations, segmentation of soil and growing material, identification of crop rows and finally segmentation between weed and crop. It is good if the applicant feels comfortable with mathematics and images. But, this work can also be interested for applicants that like environmental questions, agriculture and food science.

The diploma work will be supervised by Mats Rudemo at Mathematical Statistics and Niklas Lorén at SIK, and it forms a part of a large SLF project about site specific control of the herbicide dose. If you are interested please contact Mats Rudemo, rudemo@math.chalmers.se, 031 ¿ 772 35 75 or Niklas Lorén, nl@sik.se, 031 ¿ 335 13 14 for further information.


Informationen om uppsatsförslag är hämtad från Nationella Exjobb-poolen.