Image Processing for Plant Detection: Difference between revisions
(Created page with "= Description = In a research cooperation of SCC with da-cons, we are investigating methods and algorithms to detect a plant on a photograph. The overall aim is to calulate, ...") |
No edit summary |
||
| (2 intermediate revisions by the same user not shown) | |||
| Line 1: | Line 1: | ||
= Description = |
= Description = |
||
In a research cooperation of SCC with da-cons, we are investigating |
In a research cooperation of SCC with da-cons, we are investigating methods and algorithms to detect a plant on a photograph. The overall aim is to calulate, e.g., the age or condition of the plant. Derived from this, we offer an enclosed task for students to detect if the pixel of an image is part of a plant or not. |
||
methods and algorithms to |
|||
detect a plant on a photograph. The overall aim is to calulate, e.g., |
|||
the age or |
|||
condition of the plant. Derived from this, we offer an enclosed task for |
|||
students to detect |
|||
if the pixel of an image is part of a plant or not. |
|||
= Tasks = |
= Tasks = |
||
To achieve the described goals, you need to setup a toolchain first to exclude as many irrelvant information as possible from the given images. After identifying the plants in the bed, you need to develop a filter to only recognize the pixels of the plants in it. Note, that the angle of the image might change, i.e., your method needs to be robust and able to cope with such images. As an extension, we also have images that were shot with an infrared camera during the night which make the distinction more complex. After implementing your approach, you need to evaluate and write a documentation (including theoretical aspects and your approach, ~ x pages) about it. |
|||
To achieve the described goals, you need to setup a toolchain first to |
|||
exclude as many irrelvant information |
|||
as possible from the given images. After identifying the plants in the |
|||
bed, you need to develop a filter to only |
|||
recognize the pixels of the plants in it. Note, that the angle of the |
|||
image might change, i.e., your method needs to |
|||
be robust and able to cope with such images. As an extension, we also |
|||
have images that were shot with an infrared camera during the night |
|||
which make the distinction more complex. After implementing your |
|||
approach, you need to evaluate and write a documentation |
|||
(including theoretical aspects and your approach, ~ x pages) about it. |
|||
After the project has finished, you also have to give a presentation (~ |
After the project has finished, you also have to give a presentation (~ 30 mins) about your achievements. |
||
x mins) about your achievements. |
|||
= Requirements = |
= Requirements = |
||
* familiarity with C/C++ |
* familiarity with C/C++ |
||
* knowlegde/understanding of image processing, ideally experience with |
* knowlegde/understanding of image processing, ideally experience with an image processing library, e.g., OpenCV [0] |
||
an image processing library, e.g., OpenCV |
|||
= References = |
= References = |
||
: [0] [ |
: [0] [http://opencv.org/ http://opencv.org] |
||
= Contact = |
= Contact = |
||
Latest revision as of 19:15, 12 September 2016
Description
In a research cooperation of SCC with da-cons, we are investigating methods and algorithms to detect a plant on a photograph. The overall aim is to calulate, e.g., the age or condition of the plant. Derived from this, we offer an enclosed task for students to detect if the pixel of an image is part of a plant or not.
Tasks
To achieve the described goals, you need to setup a toolchain first to exclude as many irrelvant information as possible from the given images. After identifying the plants in the bed, you need to develop a filter to only recognize the pixels of the plants in it. Note, that the angle of the image might change, i.e., your method needs to be robust and able to cope with such images. As an extension, we also have images that were shot with an infrared camera during the night which make the distinction more complex. After implementing your approach, you need to evaluate and write a documentation (including theoretical aspects and your approach, ~ x pages) about it.
After the project has finished, you also have to give a presentation (~ 30 mins) about your achievements.
Requirements
- familiarity with C/C++
- knowlegde/understanding of image processing, ideally experience with an image processing library, e.g., OpenCV [0]
