SEMINAR REPORT

ON

FISH SPECIES AUTOMATED DETECTION SYSTEM

(A CASE STUDY OF BAKOLORI DAM)

By

MOMOH OYIZA RASHEEDA

1707231007
AND
ABDULRAHMAN MAIMUNA
1707231008
A SEMINAR PRESENTATION SUBMITTED TO THE DEPARTMENT OF
COMPUTER SCIENCE , SCHOOL OF SCIENCE AND TECHNOLOGY,
ABDU GUSAU POLYTECHNIC TALATA MAFARA ZAMFARA STATE

IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR NATIONAL

DIPLOMA (ND) IN COMPUTER SCIECNE .

OCTOBER, 2019.

1
 APPROVAL PAGE

This is to certify that the Seminar research work was carried out by MOMOH OYIZA
RASHEEDA 1707231007 AND ABDULRAHMAN MAIMUNA 1707231008
, in partial fulfilment of the requirement for the award of National Diploma (HND) in
Department of computer science, Abdu Gusau Polytechnic Talata Mafara, Zamfara state.

________________________ ________________
Dr. Samaila Musa Date/Sign
Seminar Supervisor

2
 ABSTRACT
The fish species automatic detection system is a system that can detect, at the same time
distinguishes the various species of fishes using images. The system offers a
comprehensive knowledge in detecting, recognizing and distinguishing the species of fish
by first storing the images of fishes in the database, make the system to have a
compatibility of receiving the image from your computer and then compare it with the
existing images in the database, if match appear with the one of the images in the
database, the system will also provide a sub menu to the user to add the fish with its
descriptions. Method used are primary and secondary method The system also has a
room to make some modifications or corrections in the database, in terms of the data, the
images of the fishes, to delete the database of specific fish or the entire database in
general. This program is been developed using MYSQL and Java programming language
for it user friendly.

3
INTRODUCTION

The fish species automatic detection system is a system that can detect, at the same time
distinguishes the various species of fishes. The system offers a comprehensive
knowledge in detecting, recognizing and distinguishing the species of fish by first storing
the images of fishes in the database, make the system to have a compatibility of receiving
the image from your computer and then compare it with the existing images in the
database, if match appear with the one of the images in the database, then the system
will return that image with some descriptions that prove the fish else return the statement
that show the invalidity of that image. The system also has a room to make some
modifications or corrections in the database, in terms of the data, the images of the fishes,
to delete the database of specific fish or the entire database in general.
STATEMENT OF THE PROBLEMS
The classification and distinguishing of the fish species encounter a numerous problems
which include the following:

a. Different authors mistakenly used different name for the same species.
b. The gill rakers are varying in number and size with different species and also
with the age in appearance.
c. Some species are difficult to identify at early stage.
d. Length of a distinct fish is also difficult to measure.
e. Color of some fish.
f. The size and the shape of the fin of a fish are often as a clue to the identity of
a species.
RESEARCH QUESTIONS
The study is poised towards providing answers to the following research questions
 How can you identify a fish species ?
 Can a gill rakers be identified in number and size with different species?
 How can some species be identify at this early stage?
 Can the length of a distinct fish be measured?
 How can some fish be identify by their colors ?
 Can the size and the shape of the fin of a fish be identify?

4
SIGNIFICANCES OF THE STUDY
 The aim of this work is to perform a comprehensive review of existing methods for
object detection and recognition suitable for passing fish, to select the most
promising ones,
 To design a system that will be able collects high-resolution images of fish behavior
in all riverine conditions—light, dark, murky.
 To design a system that will be able monitor numerous fish of many life stages
across a great distance
 To develop a system for detecting fish in underwater
 To implement an application that is able to detect fish in the video, distinguish
fish from other passing objects and extract features useful for fish
AIM AND OBJECTIVES OF THE STUDY
The aim of the study is to design an automatic system that can identify classify and
recognize the species of fish by images in order to solved or approximately reduce the
numerous problems faced by manual system as well as the some of the automatic
systems.
The objectives of automatic detection of fish species system is mention below:
a. To design and implement the system that can recognize and detect the species
of fish efficiently and effectively using images.
b. To reduce the detail delay and the time spent by manual system.
c. To solve the problem of color, length and shape of fishes, faced by fishery.
d. To simplify the most of the automatic systems by making the propose system
simple (readable, understandable and supportable).
THE SCOPE AND LIMITATION OF THE STUDY
The scope of this research is to design differentiate types of detection systems but
the researchers confined this study to design a fish species detection system that
captures collect and store the images and the details of twenty species of fishes in the
database which will enables the end users to search the existing image of distinct species
and its description respectively and the system is limited with only twenty species of fishes
that is, the system can only detect, recognize and analyze the images of twenty species
of fishes. This is due to the time factor, availability of materials and financial constraints.

5
LITERATURE REVIEW

INTRODUCTION

In this chapter the researchers are going to review the related literatures and shed more
light on some vital concept related to the research topics. Each family of fishes has
physical traits that set it apart from others, called distinguishing characteristics. These
characteristics help fish to survive in their environment. By observing and comparing
these features, students learn that fish, like other living organisms, can be organized and
classified into meaningful groups for identification and further study. With the exception of
some primitive species, most fish have common characteristics that include gills, scales,
fins, and bony skeletons. Some characteristics that differentiate fish include head shape
and mouth orientation, fin type and location, and average adult size. Color markings, such
as vertical stripes or fin spots, may also help differentiate fish when used in combination
with other factors including geographic range. Distinguishing characteristics can provide
clues about where a species typically lives and what it eats. For example, fish in the
sturgeon and sucker families have downward oriented mouths (sometimes called ventral)
that enable them to find food along a lake or stream bottom. Other traits such as fin shape
and location can provide clues about whether a fish is generally a fast swimmer or a slow
swimmer. Switzer, A. (2007). Great lakes identifications.

FISH AND ITS CLASSIFICATION

Fish are poikilothermic or cold blooded animals that live in the aquatic
environment. Most fish, especially the recent species have scales on their body and
survive in the aquatic environment by their gills for respiration. Another major
characteristic of a typical fish is the present of the operculum which covers the posterior.
Ichthyology is the scientific study of fish (introduction to fisheries management). There
are about 28,100 species of fishes known to science, there are divided into four classes
59 orders, 490 families and 4,300 or so genera. The classification system of is not
stagnant and will change when we have many knowledge (fishes of the world third edition
by Joseph S. Nelson 1994).

“Nelson (2006) estimates the total number of species of fish as 32,500. Of these, some
28,400 are considered valid species, where a valid species is one that consists of groups

6
interbreeding populations that are productively isolated from other taxa (Nelson, 1999).
“It is likely that the numbers of valid species will increase as candidates are more
thoroughly documented”. (Nelson, 2006).

CONCEPT OF FISH DETECTION SYSTEM

The identification of different species of fish has become an important concern for
recreational fishermen. The proliferation of regulations relating to minimum and
possession limits compels fishermen to make the proper identification of every fish
caught. The system used to identify the species can be broadly classified into two
respectively:

i. Manual system and

ii. Automatic system

ANALYSIS OF EXISTING SYSTEM PROCEDURE

The procedure used to describe the species of fish for the existing system can be based
on manual without any key, manual system with a key or automatic system. For manual
system without any key the procedure was completely or approximately base on the
human experience which socially require an individual to have certain knowledge on fish
for him to present a good result. The manual system require human to have an experience
on fish classification, fish characteristics, fish taxonomy, physical and environmental
behaviors of fish, for him to be able to recognized a fish. The name (both author and
scientific), color, scale, fin, length of fish also help reasonably. After having all mention
above then it require you to look on a certain fish physically or on images and judge upon
him., but for the manual system that make the use of key require you to look on the image
and search for the match information in the key, while for the automatic system the
procedure used to identify the species of fish without human intervention need to be
designed by the programmers but most of them make the of images.

The collections of requirement needed for existing system to work successfully are
mentions below:

 Detail knowledge about fish descriptions, fish classification, fish morphology, and
fish taxonomy, the images of fish which will be require for basic descriptions of the

7
 most species fish by ichthyologist, knowledge about some non-stagnant feature of
fish such as color, fin, head shape, and mouth orientation and so on.
 Dichotomous key for manual system that makes the use of the key.

Images.

RESEARCH METHODOLOGY

PRIMARY DATA
This is one of the sources in which a research gets direct account, which is obtainable
from observation, questionnaire, etc. The primary data being the first hand information
obtained as far as this research is concerned. The researcher used the following ways;

Personal Interview Method:

This technique collects factors by interviewing personal connected with the system under
investigation. It is considered that they posses vital information that relating to the system
with which they are concerned. Interview was conducted through the management on
what strategy they usually considered in setting up their motivation system.

Questionnaire:
The questionnaire is a set of printed question which research employee in an attempt to
elicit the vital information from the staff. Question could be open ended. It has the
advantage of constraining pre-formulate questions and answers to which are essential
for the development of the system under consideration. The structure type of
questionnaire was chosen as the main tools for the collection of data.
SECONDARY DATA
This is the available information on the subject matter from textbooks, magazines,
newspapers, and so on. In this research, the existing literature was intensively from
textbooks, reports, presentation, internet and articles on the topic and these form the
secondary data.

SYSTEM IMPLEMENTATION

After having the user acceptance of the new system developed, the
implementation phase begins. Implementation is the stage of a project during which

8
theory is turned into practice. The major steps involved in this phase are:

 The hardware and the relevant software required for running the system must be
made fully operational before implementation.
 The conversion is also one of the most critical and expensive activities in the
system development life cycle.
 The data from the old system needs to be converted to operate in the new format
of the new system.
 The database needs to be setup with security and recovery procedures fully
defined. During this phase, all the programs of the system are loaded onto the
user’s computer. After loading the system, training of the user starts.
 RESULT/FINDINGS
 The fish species detection System was presented to different users with the intent of
finding errors and observing weather it behaves as expected. The faults were corrected
and the process was repeated until the system was proven to be working according to
users’ specification and performance requirements.
 The system was also tested to see whether it was capturing invalid data, this was done
by putting wrong data and then the system responded by alert messages displaying the
type of error. Testing and validation was done successfully.
SUMMARY
 The System was presented to different users so as to get feedback about the system
performance as to whether the system met their needs or user requirements for which it
was designed for. The process involved checking input and output data of the system to
ensure that they are complete and accurate especially in the area of database to check
whether the system conformed to the standards of similar systems under defined operating
conditions. Further tests on validation were carried out on the system to verify that it met
the specified user requirements.
 The users were satisfied with the system and concluded that the system was simple to
use allowing them to navigate through the system with ease. The system was fast in
responding to the different requests and that it satisfied the intended user needs or

9
 requirements. A questionnaire was also designed to capture their responses and
thoughts.

CONCLUSION

The fish species automated detection system have been implemented with the
great achievement of been able to search an image in the system together with a full
descriptions of that image been searching in the system.

RECOMMENDATION

The fish species automated detection system has a lot of recommendations but some of
them are mentions below:

 The accuracy of the system from the result of the testing fails to be completely
automatic but rather than base on the database created by the admin because
none of the classifier work successfully.
 The system fails to understand different images of the same species.
 The system fails to provide facility that the users or experts will use to make a
comment base on what you see about the system.
 The system suffers from a lot of noise from most of the images.
 The system fails to recognize the same image of different format.
 The system lack security.

10
 REFERENCES

Anna Switzer, 2007 Identifying of the great lakes fish FLOW Unit 3: FISH | MICHU 08-
403 | © Michigan Sea Grant, Regents of the University of Michigan |
www.projectflow.us look one of the following www.projectfl ow.us
http://www.projectflow.u.
D. malik 2010 JAVA Programming-From Prob. Analysis to Pgm. Design 4th ed.-
(Cengage) BBS.pdf.

Deitel P. J. 2012, Deitel H. M.-Java How to program, 9th Edition.pdf.

Dr. Jawahar Vetter: Prof. Dharminder KumarOverview of System Analysis & Design.

Hernandez–serna A., Jemenez–Segura, L.F. (2014) Automatic identification of species

with neural networks.

I. Magawata. 2012 Fish of northern Nigeria.

Jabaka 2012 System modeling and simulation lecture note.

John G. Casey Fish identification guide. Funded by salt water recreational fishing
development fund.

Joseph, S. N. (1994) Fishes of the world (3rd Edition).

Malan Sifiyanu, colony. (personal interview, September 8, 2015) Bakolori Dam history and
irrigation project.

Method of fish biology, publisher: American Fisheries Society, Editors: Moyle, P., Schreck,
P. pp. 109 – 140.

11