In this article, the evolution of BFT (bluefin tuna) sizes in fattening cages is studied, for which it was necessary to perform exhaustive monitoring with stereoscopic cameras and an exhaustive analysis of the data using automatic procedures. Exploring the size evolution of BFT over a long period is an important step in inferring their growth patterns, which are essential for designing smart aquaculture and sustainable fishing, and even assessing their health status. An important objective of this work was to verify whether tuna in captivity, in addition to fattening, grow in length. To this end, our autonomous monitoring system, equipped with stereoscopic cameras, was installed from 28 July 2020 to 23 May 2021 in a fattening cage in the Mediterranean containing 724 free-swimming tuna. This system provides thousands of images that, grouped by time intervals, allow us to conduct our studies. An automatic procedure, already introduced in a previous work and capable of processing large volumes of data, is used to estimate the length and width of individuals in ventral stereoscopic images of fish, and the evolution over time is analysed for each biometric characteristic. However, verifying the evolution of length and width based only on means or medians of these measurements may be inconsistent and insufficiently accurate to support our study objectives, as individuals of different sizes and ages may grow at different rates. Therefore, a modal analysis (Bhattacharya’s method) was undertaken to identify the cohorts within the population. The results showed that each modal length surpassed the length of the next cohort and that there was accelerated growth in cages compared to the wild. In addition, we proved that using a length–width–weight relationship to estimate fish weight gives more accurate results than traditional length–weight relationships for fish fattened in cages.

The classification of honey pollen grains is performed in order to classify honey according to its botanical origin, which is of great importance in terms of marketing. This visual work is currently done by human specialists counting and classifying the pollen grains in microscopic images. This is a hard, time-consuming, and subject to observer variability task. Thus, automated methods are required to overcome the limitations of the conventional procedure. This paper deals with the automatic classification of honey pollens using five representative Neural Networks coming from the ImageNet Challenge: VGG16, VGG19, ResNet50, InceptionV3 and Xception. The ground truth is composed of 9983 samples of 16 different types of pollens corresponding to citrus and rosemary pollens and its companions. The best result was obtained with the InceptionV3 network, achieving an accuracy of 98.15%, that outperforms the results obtained in previous works.

This paper proposes an indirect method to estimate Bluefin Tuna (Thunnus thynnus) biomass in cages using acoustic techniques. Two Simrad EK60 echosounders working at 120 and 200 kHz and a stereo camera were used to obtain target strength (TS) to fork length (FL) relationships for both operating frequencies. The equipment was placed at the bottom of a floating cage, facing towards the surface to record the ventral aspect of fish. The acoustic and optical recordings were automatically analysed and the combination of acoustic and optical results provided unequivocal TS-FL assignments. Good relationships between TS and FL were obtained for both frequencies even without discriminating data from different fish tilts and without using beam directivity compensation. Stronger correlations were obtained for compensated TS at 200 kHz when reduced tuna swimming tilt was considered. TS measurements were compared to MFS numerical predictions for a Bluefin swimbladder model, with the simulation results showing good agreement with experimental measurements. The results allow the mean tuna length to be predicted in growing or fattening cages from acoustic data raising the possibility of improving production management and of providing a useful tool for catch control estimations made by international organisations (like ICCAT) and government bodies. (c) 2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

A proposal is described for an underwater sensor combining an acoustic device with an optical one to automatically size juvenile bluefin tuna from a ventral perspective. Acoustic and optical information is acquired when the tuna are swimming freely and the fish cross our combined sensor's field of view. Image processing techniques are used to identify and classify fish traces in acoustic data (echogram), while the video frames are processed by fitting a deformable model of the fishes' ventral silhouette. Finally, the fish are sized combining the processed acoustic and optical data, once the correspondence between the two kinds of data is verified. The proposed system is able to automatically give accurate measurements of the tuna's Snout-Fork Length (SFL) and width. In comparison with our previously validated automatic sizing procedure with stereoscopic vision, this proposal improves the samples per hour of computing time by 7.2 times in a tank with 77 juveniles of Atlantic bluefin tuna (Thunnus thynnus), without compromising the accuracy of the measurements. This work validates the procedure for combining acoustic and optical data for fish sizing and is the first step towards an embedded sensor, whose electronics and processing capabilities should be optimized to be autonomous in terms of the power supply and to enable real-time processing.

This paper presents a non-invasive fully automatic procedure to obtain highly accurate fish length estimation in adult Bluefin Tuna, based on a stereoscopic vision system and a deformable model of the fish ventral silhouette. The present work takes a geometric tuna model, which was previously developed by the same authors to discriminate fish in 2D images, and proposes new models to enhance the capabilities of the automatic procedure, from fish discrimination to accurate 3D length estimation. Fish length information is an important indicator of the health of wild fish stocks and for predicting biomass using length-weight relations. The proposal pays special attention to parts of the fish silhouette that have special relevance for accurate length estimation. The models have been designed to best fit the rear part of the fish, in particular the caudal peduncle, and a width parameter has been added to better fit the silhouette. Moreover, algorithms have been developed to extract snout tip and caudal peduncle features, allowing better initialization of model parameters. Snout Fork Length (SFL) measurements using the different models are extracted from images recorded with a stereoscopic vision system in a sea cage containing 312 adult Atlantic Bluefin Tuna. The automatic measurements are compared with two ground truths: one configured with semiautomatic measurements of favourable selected samples and one with real SFL measurements of the tuna stock collected at harvesting. Comparison with the semiautomatic measurements demonstrates that the combination of improved geometric models and feature extraction algorithms delivers good results in terms of fish length estimation error (up to 90% of the samples bounded in a 3% error margin) and number of automatic measurements (up to 950 samples out of 1000). When compared with real SFL measurements of the tuna stock, the system provides a high number of automatic detections (up to 6706 in a video of 135 min duration, i.e., 50 automatic measurements per minute of recording) and highly accurate length measurements, obtaining no statistically significant difference between automatic and real SFL frequency distributions. This procedure could be extended to other species to assess the size distribution of stocks, as discussed in the paper.

This article presents a non-invasive fully automatic procedure for Bluefin Tuna sizing, based on a stereoscopic vision system and a deformable model of the fish ventral silhouette. An image processing procedure is performed on each video frame to extract individual fish, followed by a fitting procedure to adjust the fish model to the extracted targets, adapting it to the bending movements of the fish. The proposed system is able to give accurate measurements of tuna snout fork length (SFL) and widths at five predefined silhouette points without manual intervention. In this work, the system is used to study size evolution in adult Atlantic Bluefin Tuna (Thunnus Thynnus) over time in a growing farm. The dataset is composed of 12 pairs of videos, which were acquired once a month in 2015, between July and October, in three grow-out cages of tuna aquaculture facilities on the west Mediterranean coast. Each grow out cage contains between 300 and 650 fish on an approximate volume of 20 000 m(3). Measurements were automatically obtained for the 4 consecutive months after caging and suggest a fattening process: SFL shows an increase of just a few centimetres (2%) while the maximum width (A(1)) shows a relative increase of more than 20%, mostly in the first 2 months in farm. Moreover, a linear relation (with coefficient of determination R-2 > 0.98) between SFL and widths for each month is deduced, and a fattening factor (F) is introduced. The validity of the measurements is proved by comparing 15 780 SFL measurements, obtained with our automatic system in the last month, versus ground truth data of a high percentage of the stock under study (1143 out of 1579), obtaining no statistically significant difference. This procedure could be extended to other species to assess the size distribution of stocks, as discussed in the article.

This paper proposes a robust deformable adaptive 2D model, based on computer vision methods, that automatically fits the body (ventral silhouette) of Bluefin tuna while swimming. Our model (without human intervention) adjusts to fish shape and size, obtaining fish orientation, bending to fit their flexion motion and has proved robust enough to overcome possible segmentation inaccuracies. Once the model has been successfully fitted to the fish it can ensure that the detected object is a tuna and not parts of fish or other objects. Automatic requirements of the fishing industry like biometric measurement, specimen counting or catch biomass estimation could then be addressed using a stereoscopic system and meaningful information extracted from our model. We also introduce a fitting procedure based on a fitting parameter – Fitting Error Index (FEI) – which permits us to know the quality of the results. In the experiments our model has achieved very high success rates (up to 90%) discriminating individuals in highly complex images acquired for us in real conditions in the Mediterranean Sea. Conclusions and future improvements to the proposed model are also discussed.

This paper presents an application of visual quality control in orange post-harvesting comparing two different approaches. These approaches correspond to two very different methodologies released in the area of Computer Vision. The first approach is based on Multivariate Image Analysis (MIA) and was originally developed for the detection of defects in random color textures. It uses Principal Component Analysis and the T2 statistic to map the defective areas. The second approach is based on Graph Image Segmentation (GIS). It is an efficient segmentation algorithm that uses a graph-based representation of the image and a predicate to measure the evidence of boundaries between adjacent regions. While the MIA approach performs novelty detection on defects using a trained model of sound color textures, the GIS approach is strictly an unsupervised method with no training required on sound or defective areas. Both methods are compared through experimental work performed on a ground truth of 120 samples of citrus coming from four different cultivars. Although the GIS approach is faster and achieves better results in defect detection, the MIA method provides less false detections and does not need to use the hypothesis that the bigger area in samples always correspond to the non-damaged area.

In this paper we introduce an algorithm aimed to create a background model with multimodal support, which associates a confidence value to the obtained model. Our algorithm creates the model based on a criterion of motion, pixel behavior and pixel similarity with the scenes background. This method uses only three frames to create a first model without restrictions on the frame content. The model is adapted over time to reflect new situations and illumination changes in the scene. One approach to detect corrupt model is also mentioned. The goal of confidence value is to quantify the quality of the model after a number of frames have been used to build it. Quantitative experimental results are obtained with a well-known benchmark and compared to a classical background modelling algorithm, showing the benefits of our approach.

Background subtraction is a popular technique in video surveillance. In order to use it, a background model must be built and updated according to light and scenario changes. We discuss in this paper a new algorithm (BAC) which creates or restores a back-ground model based on the behaviour of pixels in successive frames, while performs a segmentation of objects in the scene yielding a confidence value for the obtained background, a problem which is addressed by few methods in the literature. This allows us to fulfil the requirement of producing a model, for instance in scenarios like airport halls, without interfering normal operation and still segment scenes.

This paper describes an approach to classify people, groups of people and luggage in the halls of an airport. The algorithm is included into a surveillance system which tracks and classifies objects and transmits this information to a higher computational level which fuses the information of several cameras covering overlapping areas. Two kind of features are used: foreground density features and features related to real-size of objects, obtained by applying a homographic model. A classification schema based on k-nn classifiers and a voting system makes the classification process highly robust. On-line and off-line experiments are introduced.