Study on leave image processing with application in herbal classification and early detection of chili plant disease
Herbs have been widely used in food preparation, medicine and cosmetic industry. Knowing which herbs to be used would be very important in these applications. The current way of identification and determination of the types of herbs however, is still being done manually and prone to human error....
Saved in:
| Format: | Thesis |
|---|---|
| Language: | English |
| Subjects: | |
| Online Access: | http://dspace.unimap.edu.my:80/xmlui/bitstream/123456789/77179/1/Page%201-24.pdf http://dspace.unimap.edu.my:80/xmlui/bitstream/123456789/77179/2/Full%20text.pdf http://dspace.unimap.edu.my:80/xmlui/bitstream/123456789/77179/4/Zulkifli%20Husin.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Herbs have been widely used in food preparation, medicine and cosmetic industry. Knowing
which herbs to be used would be very important in these applications. The current way of
identification and determination of the types of herbs however, is still being done manually
and prone to human error. Designing a convenient and automatic recognition system of herbs
species is essential since this will improve herb species classification efficiency. Chili
(Capsicum Annum and Capsicum Frutescen) is an important fruiting vegetable used in
majority of Asian dishes. Chili cultivation has been a very difficult and meticulous task due
to its vulnerability to various attacks frommicro-organisms, bacterial disease and pests which
leave distinguished marks on leaves, stems or fruits. Current manual method applies
pesticides and chemicals indiscriminately throughout the farm. To improve the process,
development of an automated disease detection is essential. There are a few research that
have been done in classification of the plant species using certain factors (leaf shape and
size). The classification are accomplished through several image processing techniques.
However, the literature shows that there are still a gap in classifying the herb plants species.
Therefore, this research focuses on classification approach to the shape, texture features and
colors of the herbs leaves. The combination of techniques used in morphology image
processing i.e. SVD and skeleton would be able to classify the species of herb regardless of
the shape and size. In addition, the techniques demonstrate the capability to detect early plant
chili disease through leaf features inspection using HSV colour model technique. The
proposed herbs species recognition system employs neural networks algorithm and image
processing techniques to perform classification on twenty herbs species. One hundred
samples for each species went through the system and the recognition accuracy was at 98.9%.
Most importantly the system is capable of identifying the herbs leaves species even though
they are dried, wet, torn or deformed. Additionally, a novel method of early automatic
recognition for plant chili disease based on color and texture features using a HSV color
model and BPNN technique via intelligent decision support system is presented in this
research. The proposed system employs image processing technique on one thousand chili
plant samples and the recognition accuracy was at 97.7%. The efficiency and effectiveness
of the proposed methods in recognizing herbs plant and detecting early plant chili disease are
demonstrated by the experiments. |
|---|