A new image dataset for the evaluation of automatic fingerlings counting

Garcia, Vanir; Sant’Ana, Diego André; Zanoni, Vanda Alice Garcia; Pache, Marcio Carneiro Brito; Naka, Marco Hiroshi; Albuquerque, Pedro Lucas França; Lewandowski, Tiago; Oliveira, Adair da Silva; Rozales, João Victor Araújo; Ferreira, Milena Wolff; Queiroz, Eduardo Quirino Arguelho de; Almanza, José Carlos Marino; Pistori, Hemerson

doi:10.1016/j.aquaeng.2020.102064

Cited by 15 publications

(6 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Garcia et al [13] have already published preliminary results on the evaluation of the fingerling counter through its exploration of parameters. The best performance was obtained with an MSE of 2.65 and R squared 0.9803.…”

Section: Related Workmentioning

confidence: 99%

“…The equipment responsible for the fingerling counting, was developed in [13], and basically consists of 4 main components: the structure, the lighting, the camera and the software. The structure is assembled in a way that it is inclined, around 11 to 13 degrees, through which the fingerlings slide with the help of water, which flows continuously.…”

Section: A Fingerling Countermentioning

confidence: 99%

“…Equipment with embedded computer vision-based software tools have been studied and developed in order to support this counting process [7]- [11]. Among this equipment and software reported, we highlight the fingerling counter developed by Inovisão group, whose main requirement refers to the counting of fingerlings of the species Pintado Real (Pseudoplatystoma corruscans) [12], [13].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

An Investigation of Parameter Optimization in Fingerling Counting Problems

Oliveira¹,

Pache²,

Rezende

et al. 2021

Anais Do XVII Workshop De Visão Computacional (WVC 2021)

Self Cite

View full text Add to dashboard Cite

The objective of this paper is to investigate which combination of parameters for the fingerling counting software results in the smallest Mean Absolute Error (MAE) and smallest Root Mean Squared Error (RMSE). For this, an image dataset called FISHCV155V was created and separated into training and test sets, where different combinations of parameters for the software were tested. From the obtained results were extracted individual performance metrics for each combination of parameters, such as MAE, Mean Square Error (MSE) and RMSE. Video frames were analysed comparing the parameter combination that obtained the best and worst results, in order to investigate the influence of such parameters in the performance of the software. From such results, it was concluded that the best combination reached 5.99 MAE and 9.96 RMSE.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: A Fingerling Countermentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Investigation of Parameter Optimization in Fingerling Counting Problems

Oliveira¹,

Pache²,

Rezende

et al. 2021

Anais Do XVII Workshop De Visão Computacional (WVC 2021)

Self Cite

View full text Add to dashboard Cite

show abstract

“…In 2019, Albuquerque et al proposed an automatic counting system [6] combining point detection, Gaussian mixture, and Kalman filter, which used the background difference method to detect fry, and the Kalman filter to reduce the error due to fish overlap. They also proposed a test dataset in 2020 [7]. Due to the excellent performance of the Convolutional Neural Network (CNN) in object detection and tracking, in 2019, Lainez and Gonzales [8] tested the effectiveness of CNN-based in fish detection and counting accuracy on tilapia of four sizes and set different thresholds for different numbers of fish species present.…”

Section: Introductionmentioning

confidence: 99%

Automatic Counting Method of Fry Based on Computer Vision

Juan

Sun

Cui

et al. 2023

IEEJ Transactions Elec Engng

View full text Add to dashboard Cite

Nowadays, fish farming plays an essential role in aquaculture. In the process of fry sales, the traditional manual counting method takes a great deal of time. Because the fish are constantly moving in the water, they are difficult to count and are likely to be unnecessarily damaged. With the rise of computer vision, fry counting based on image processing is also developing. It has the advantages of having high accuracy and fast processing speed. The primary counting process is divided into detection, adhesion segmentation, and fry counting. First, we improved the maximum interclass variance threshold selection method, and we extracted the histogram features under different channels to detect the fry; then, a matching segmentation method based on concave and skeleton crossing points was used to segment the adherent fry to improve the counting accuracy, the concaves were detected by CNN. We tested the method on real fry data sets, and the average accuracy reached 97%. Our approach had better accuracy than any existing current fish counting methods. © 2023 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.

show abstract

“…Traditional machine learning-based counting methods use techniques such as image segmentation, which requires the human extraction of features, manual setting of thresholds, and setting of regression functions to perform counting. Researchers like Ibrahin et al [18], Albuquerque et al [19], Zhang et al [20], and Garcia et al [21] adopted traditional computer vision processing methods such as speckle detection and edge contour extraction to achieve the automatic counting of fry. Researchers such as Zhang et al [22] further utilized binarization and expansion erosion to determine the biomass of fry through independently refined connected domains, improving the accuracy of overlapping regions to some extent.…”

Section: Introductionmentioning

confidence: 99%

Fry Counting Models Based on Attention Mechanism and YOLOv4-Tiny

et al. 2022

View full text Add to dashboard Cite

Accurate counting is difficult in the case of large numbers of overlapping and adhering fry. In this study, we propose a lightweight target detection counting method based on deep learning methods that can meet the deployment requirements of edge computing device for automatic fry counting while obtaining a high counting accuracy. We improve the structure of YOLOv4-tiny by embedding different attention mechanisms in the CSP blocks of the backbone network to enhance the feature extraction performance. In addition, the low efficiency of feature fusion in the original model is also addressed by adding different attention mechanisms to the neck network structure to promote the effective fusion of deep feature information with shallow feature information and improve the counting accuracy. The experimental results showed that the six models proposed in this study improved the model accuracy and recall to varying degrees compared with the original YOLOv4-tiny model, while retaining the advantages of the YOLOv4-tiny model in terms of its small number of parameters and fast inference rate. It was also shown that the CBAM(n)-YOLOv4-tiny model obtained by adding the CBAM to the neck network showed the most significant improvement, with an mean average precision (mAP) of 94.45% and a recall of 93.93%.

show abstract

A new image dataset for the evaluation of automatic fingerlings counting

Cited by 15 publications

References 7 publications

An Investigation of Parameter Optimization in Fingerling Counting Problems

An Investigation of Parameter Optimization in Fingerling Counting Problems

Automatic Counting Method of Fry Based on Computer Vision

Fry Counting Models Based on Attention Mechanism and YOLOv4-Tiny

Contact Info

Product

Resources

About