Fish population estimation and species classification from underwater video sequences using blob counting and shape analysis

2013 ◽  
Author(s):  
J. N. Fabic ◽  
I. E. Turla ◽  
J. A. Capacillo ◽  
L. T. David ◽  
P. C. Naval
Keyword(s):  
Shape Analysis ◽  
Fish Population ◽  
Population Estimation ◽  
Video Sequences ◽  
Underwater Video ◽  
Species Classification

Progress in the Automated Identification, Measurement, and Counting of Fish in Underwater Image Sequences

2016 ◽  
Vol 50 (1) ◽  
pp. 4-16 ◽  
Author(s):  
Mark R. Shortis ◽  
Mehdi Ravanbakhsh ◽  
Faisal Shafait ◽  
Ajmal Mian
Keyword(s):  
Fork Length ◽  
Length Measurement ◽  
Video Sequences ◽  
Biomass Estimation ◽  
Underwater Video ◽  
Automated Identification ◽  
Species Classification ◽  
Underwater Image ◽  
Video Systems ◽  
Population Counts

AbstractUnderwater video systems are widely used for counting and measuring fish in aquaculture, fisheries, and conservation management. To determine population counts, spatial or temporal frequencies, and age or weight distributions, snout to tail fork length measurements are performed in video sequences, most commonly using a point and click process by a human operator. Current research aims to automate the identification, measurement, and counting of fish in order to improve the efficiency of population counts or biomass estimates. A fully automated process requires the detection and isolation of candidates for measurement, followed by the snout to tail fork length measurement, species classification, as well as the counting and tracking of fish. This paper reviews the algorithms used for the detection, identification, measurement, counting, and tracking of fish in underwater video sequences. The paper analyzes the most commonly used approaches, leading to an evaluation of the techniques most likely to be a comprehensive solution to the complete process of candidate detection, species identification, length measurement, and population counts for biomass estimation.

Key-Frame Extraction Method from Single-Lens Undersea Video Sequences

2013 ◽  
Vol 404 ◽  
pp. 514-519
Author(s):  
Xiu Li ◽  
Fu Xin Gao ◽  
Tian Xiang Yan ◽  
Dong Zhi Wang ◽  
Lian Sheng Chen ◽  
...  
Keyword(s):  
Extraction Method ◽  
Video Sequences ◽  
Underwater Video ◽  
Key Frame Extraction ◽  
Brightness Change ◽  
Key Frame ◽  
Noise Point ◽  
Single Lens ◽  
Shot Boundary

The process of key-frame extraction of the undersea video is different from that on the land. The effective key-frame extraction will promote research and retrieval of underwater video. In this paper, we first introduced the characteristics of the undersea video, and then proposed a new key-frame extraction method based on Sensitive Curve brightness change for single-lens undersea video sequences which measures the light shot boundary brightness change. The experiment results show that the proposed algorithm can extract key information of the undersea video quickly, and have a good performance for the noise point.

Statistical Processing of Different Sounds of Vocalizing Fish and Mammals to Estimate Their Population Size With Two Acoustic Sensors

2019 ◽  
Vol 53 (4) ◽  
pp. 68-80 ◽  
Author(s):  
Shaik Asif Hossain ◽  
Monir Hossen
Keyword(s):  
Population Size ◽  
Statistical Processing ◽  
Environmental Dna ◽  
Marine Species ◽  
Statistical Signal Processing ◽  
Fish Population ◽  
Processing Technique ◽  
Population Estimation ◽  
Human Interaction ◽  
Signal Processing Technique

AbstractFish and mammals have an enormous impact on marine ecosystems. A proper estimation of their population size is necessary, not only for their ecological values but also for commercial purposes. Most conventional techniques for estimating fish population are visual sampling techniques, the environmental DNA (eDNA) technique, minnow traps, the removal method of population estimation, and echo integration techniques, all of which are sometimes complex and costly, require human interaction, and can be harmful for marine species. In order to overcome these limitations, in this paper, a passive acoustic fishery monitoring technique is proposed as an alternative. The method is based on a statistical signal processing technique called “cross-correlation” and different types of sounds—namely, chirps, grunts, growls, clicks, and so forth—produced by fish and mammals. Our goal was not only to propose an efficient technique for fish population estimation but also to measure its performance for different fish sounds by using numerical simulations. From the analyses of simulated results, we found that the chirp sound-generating species produced better results than the other two types of sound-generating species—the grunt- and growl-generating species.

scholarly journals Identifying blue whiting (Micromesistius poutassou) stock structure in the Northeast Atlantic by otolith shape analysis

2016 ◽  
Vol 73 (9) ◽  
pp. 1363-1371 ◽  
Author(s):  
Kélig Mahe ◽  
Clémence Oudard ◽  
Tiphaine Mille ◽  
James Keating ◽  
Patricia Gonçalves ◽  
...  
Keyword(s):  
Shape Analysis ◽  
Central Area ◽  
Fish Population ◽  
Stock Structure ◽  
Linear Discriminant ◽  
Otolith Shape ◽  
Otolith Shape Analysis ◽  
Northeast Atlantic ◽  
Blue Whiting ◽  
Mature Fish

Information on stock identification and spatial stock structure provide a basis for understanding fish population dynamics and improving fisheries management. In this study, otolith shape analysis was used to study the stock structure of blue whiting (Micromesistius poutassou) in the northeast Atlantic using 1693 samples from mature fish collected between 37°N and 75°N and 20°W and 25°E. The results indicated two stocks located north and south of ICES Divisions VIa and VIb (54°5N to 60°5N, 4°W to 11°W). The central area corresponds to the spawning area west of Scotland. Sampling year effects and misclassification in the linear discriminant analysis suggested exchanges between the northern and southern stocks. The results corroborate previous studies indicating a structuring of the blue whiting stock into two stocks, with some degree of mixing in the central overlap area.

Automatic fish species classification in underwater videos: exploiting pre-trained deep neural network models to compensate for limited labelled data

2017 ◽  
Vol 75 (1) ◽  
pp. 374-389 ◽  
Author(s):  
Shoaib Ahmed Siddiqui ◽  
Ahmad Salman ◽  
Muhammad Imran Malik ◽  
Faisal Shafait ◽  
Ajmal Mian ◽  
...  
Keyword(s):  
Neural Network ◽  
Computer Vision ◽  
Test Data ◽  
Fish Species ◽  
Classification Systems ◽  
Training Data ◽  
Underwater Video ◽  
Neural Network Models ◽  
Species Classification ◽  
Video Imagery

Abstract There is a need for automatic systems that can reliably detect, track and classify fish and other marine species in underwater videos without human intervention. Conventional computer vision techniques do not perform well in underwater conditions where the background is complex and the shape and textural features of fish are subtle. Data-driven classification models like neural networks require a huge amount of labelled data, otherwise they tend to over-fit to the training data and fail on unseen test data which is not involved in training. We present a state-of-the-art computer vision method for fine-grained fish species classification based on deep learning techniques. A cross-layer pooling algorithm using a pre-trained Convolutional Neural Network as a generalized feature detector is proposed, thus avoiding the need for a large amount of training data. Classification on test data is performed by a SVM on the features computed through the proposed method, resulting in classification accuracy of 94.3% for fish species from typical underwater video imagery captured off the coast of Western Australia. This research advocates that the development of automated classification systems which can identify fish from underwater video imagery is feasible and a cost-effective alternative to manual identification by humans.

Fish population estimation: the use of rotenone to evaluate the reliability of a removal technique

1985 ◽  
Vol 16 (4) ◽  
pp. 349-357
Author(s):  
D. C. HOCKIN ◽  
K. O'HARA ◽  
D. CRAGG-HINE ◽  
J. W. EATON
Keyword(s):  
Fish Population ◽  
Population Estimation ◽  
Removal Technique

scholarly journals Robust Abnormal Event Recognition via Motion and Shape Analysis at ATM Installations

2015 ◽  
Vol 2015 ◽  
pp. 1-10 ◽  
Author(s):  
Vikas Tripathi ◽  
Durgaprasad Gangodkar ◽  
Vivek Latta ◽  
Ankush Mittal
Keyword(s):  
Support Vector Machine ◽  
Shape Analysis ◽  
Window Size ◽  
Event Recognition ◽  
Video Sequences ◽  
Support Vector ◽  
Automated Teller Machines ◽  
Average Accuracy ◽  
Hu Moments ◽  
Machine Analysis

Automated teller machines (ATM) are widely being used to carry out banking transactions and are becoming one of the necessities of everyday life. ATMs facilitate withdrawal, deposit, and transfer of money from one account to another round the clock. However, this convenience is marred by criminal activities like money snatching and attack on customers, which are increasingly affecting the security of bank customers. In this paper, we propose a video based framework that efficiently identifies abnormal activities happening at the ATM installations and generates an alarm during any untoward incidence. The proposed approach makes use of motion history image (MHI) and Hu moments to extract relevant features from video. Principle component analysis has been used to reduce the dimensionality of features and classification has been carried out by using support vector machine. Analysis has been carried out on different video sequences by varying the window size of MHI. The proposed framework is able to distinguish the normal and abnormal activities like money snatching, harm to the customer by virtue of fight, or attack on the customer with an average accuracy of 95.73%.

Sign in / Sign up

Export Citation Format

Share Document