IJARCCE
International Journal of Advanced Research in Computer and Communication Engineering
Impact Factor 7.39Vol. 11, Issue 1, January 2022
DOI 10.17148/IJARCCE.2022.11103
© IJARCCE This work is licensed under a Creative Commons Attribution 4.0 International License 11
ISSN (O) 2278-1021, ISSN (P) 2319-5940
AUTOMATED FISH FINGERLINGS
COUNTER SYSTEM: AN EVALUATION OF
IMAGE SEGMENTATION ALGORITHMS IN
OVERLAPPING OBJECTS
Nel R. Panaligan
Faculty, Computing Department, Davao del Sur State College, Digos City, Philippines
Abstract: The development of automatic fish counters had been driven by the need for accurate, long-term and cost-
effective counting and recognition for the advancement of aquaculture in the Philippines. Non-invasive methods of fish
counting are ultimately limited by the properties of the immerging technologies like when candidates for counting are
transparent and or small (Bangus Fry). Image processing is one of the most modern approaches in automating the
counting process. The main objective of the study is to evaluate three image segmentation algorithms namely (1)
Watershed Algorithm, (2) Hough Transform, (3) Concavity Analysis, in 2D image, whether or not they are capable of
segmenting two-weeks old bangus fry’s’ in an image. The basic steps involved in the conduct of this study are the
following; Image acquisition, Image Pre-Processing, Image segmentation, and Object counting. Result shows that the
first method of the Concavity Analysis which locates the contours or curve points edges of the objects in an image
performs best with the other algorithm with an accuracy rate of 94.36% with 7 false positive detections, and 154 False
Negative, in an experimental data of four sets of 2D image ranging from 100, 200, 300, and 400 bangus fry per test
image.
Keywords: Aquaculture, Image Segmentation Algorithms, Watershed Algorithm, Hough Transform, Concavity
Analysis, Evaluation
I. INTRODUCTION
Fish product contributes a significant amount of protein demand of human nutrition and its consumption have
dramatically increased - about 27 million tons of fish were consume during 1948 and this has increase to about 145
million tons during 2007 (Philippine Export 2015-2017). In a study conducted by Dowlati et al which was mentioned
by Gana et al, Fish product is about 16% of human diet all around the world (Aliyu et al., 2017). With vast growing
sector in the economy of the Philippines, fishing plays a major role in Agriculture, Hunting, Forestry and Fisheries
(AHFF) are among of the second largest contributors of the GDP at 34 and 9.2% of the economy. The Philippine
economy prides itself in having been resilient during the global financial crisis of 2008, until now, the Bangko
Sentral ng Pilipinas (BSP) says that the country has a well-insulated economy, characterized by steady remittances,
robust private consumption and capital formation, and lastly, a well-maintained services sector and a booming industrial
base.
Considering the contribution of the fishing industry in the country, it remains a question why the fish farmers
(i.e. Bangus) in the country remained in indigenous way of handling their farms, specifically in counting the fingerlings,
feeding, and classifying (i.e. size and weight). This type of farming is also laborious, especially the counting of
fingerlings to sell for interested buyers, where people find it hard to obtain the exact number of fishes because of
frequent human error, behavioural aspects of fry also affects the method of counting as workers are easily distracted and
tend to forget the total count as it accumulates later which contributes to losses, not only for fish farmers whether small
or large scale, but also for the customers.
To address the needs for a highly reliable approach necessary in the field of fisheries and aquaculture, it is vital to
develop and utilize a mostly reliable method of counting the fry in a short period of time without compromising
reliability. Artificial intelligence
This research paper will assess or evaluate three (3) image segmentation algorithms namely Watershed Algorithm,
Concavity Analysis, Morphological Transformation in bangus fry counting, where common scenario of multiple