International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 07 Issue: 01 | Jan 2020 www.irjet.net p-ISSN: 2395-0072
© 2020, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 1
SHAPE AND TEXTURE BASED INTER-SPECIFIC HYBRID FISH SPECIES
IMAGE RECOGNITION
Lakew Tarekegn
1
, Tesfa Tegegn
2
Lakew Tarekegn Berihun
UNIVERSITY OF GONDAR, COLLEGE OF INFORMATICS
Information Technology Department
Gondar, Ethiopia
BAHIR DAR UNIVERSITY, COLLEGE OF INFORMATICS
Information Technology Department
Bahir Dar, Ethiopia
Tesfa Tegegn(Phd) and Lakew T(Msc) , Dept. of Information Technology, Informatics College, State, Amhara,
Country, Ethiopia
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Recently inter-specific
hybridization is a common phenomenon in fresh
water fish’s species. Natural hybridization is
more common in fish than in other vertebrates,
and non-native fish species which have been
introduced extensive in a worldwide those are
difficult to identify manually. However, inter
specific hybridize among genetically divergent
fish are often fertile, and provide good product
that results in creation of survival fit species. In
case of this nowadays a lot of works were done
by depending on the computer in order to
processing time to be reduced when the experts
pass a key decision what type of fish species exist
in under marine area as biologist and
agricultural researcher. Due to this computer
has been used for digital image recognition
which has been extremely found and studied for
this role work to get fast response of replied
when it has been compared and contrast with
human know how knowledge level with recent
machine level knowledge one which has been
depending on computer and provide more
results which are accurate.
Even though due to hybridization the
agricultural experts has been faced a challenges
and confusion to isolate and clearly tries to
identify as well as recognize this occurring
species based on morphological shapes and
texture of fish species images. So that using these
procedural techniques the researcher work
provide attractive fish recognition based on their
types of species. The detection, recognition and
classification system is using the proposed
procedural methods. Those procedural methods
image processing approaches such as noise
removal, image enhancement, image restoration,
feature extraction, segmentation, filtering,
histogram processing, countor tracking, edge
detection, graye scale processing, Hsv color fish
image based processing. Beside this, the study
used Svm to classify and recognized those fish’s
species using the interspecific values with a
common morphological fish image shape and
texture that the fish contained. This work
scientifically contributed to recognize and
classify that those complex interbreed fish gene
type using feature extraction based on their