RESUMEN
Understanding the feeding dynamics of aquatic animals is crucial for aquaculture optimization and ecosystem management. This paper proposes a novel framework for analyzing fish feeding behavior based on a fusion of spectrogram-extracted features and deep learning architecture. Raw audio waveforms are first transformed into Log Mel Spectrograms, and a fusion of features such as the Discrete Wavelet Transform, the Gabor filter, the Local Binary Pattern, and the Laplacian High Pass Filter, followed by a well-adapted deep model, is proposed to capture crucial spectral and spectral information that can help distinguish between the various forms of fish feeding behavior. The Involutional Neural Network (INN)-based deep learning model is used for classification, achieving an accuracy of up to 97% across various temporal segments. The proposed methodology is shown to be effective in accurately classifying the feeding intensities of Oplegnathus punctatus, enabling insights pertinent to aquaculture enhancement and ecosystem management. Future work may include additional feature extraction modalities and multi-modal data integration to further our understanding and contribute towards the sustainable management of marine resources.
RESUMEN
Unsupervised domain adaptation (UDA) enables knowledge transfer from a labeled source domain to an unlabeled target domain. However, UDA performance often relies heavily on the accuracy of source domain labels, which are frequently noisy or missing in real applications. To address unreliable source labels, we propose a novel framework for extracting robust, discriminative features via iterative pseudo-labeling, queue-based clustering, and bidirectional subdomain alignment (BSA). The proposed framework begins by generating pseudo-labels for unlabeled source data and constructing codebooks via iterative clustering to obtain label-independent class centroids. Then, the proposed framework performs two main tasks: rectifying features from both domains using BSA to match subdomain distributions and enhance features; and employing a two-stage adversarial process for global feature alignment. The feature rectification is done before feature enhancement, while the global alignment is done after feature enhancement. To optimize our framework, we formulate BSA and adversarial learning as maximizing a log-likelihood function, which is implemented via the Expectation-Maximization algorithm. The proposed framework shows significant improvements compared to state-of-the-art methods on Office-31, Office-Home, and VisDA-2017 datasets, achieving average accuracies of 91.5%, 76.6%, and 87.4%, respectively. Compared to existing methods, the proposed method shows consistent superiority in unsupervised domain adaptation tasks with both fully and weakly labeled source domains.
Asunto(s)
Algoritmos , Aprendizaje Automático no Supervisado , Redes Neurales de la Computación , Análisis por Conglomerados , HumanosRESUMEN
Heterogeneous domain adaptation (HDA) aims to address the transfer learning problems where the source domain and target domain are represented by heterogeneous features. The existing HDA methods based on matrix factorization have been proven to learn transferable features effectively. However, these methods only preserve the original neighbor structure of samples in each domain and do not use the label information to explore the similarity and separability between samples. This would not eliminate the cross-domain bias of samples and may mix cross-domain samples of different classes in the common subspace, misleading the discriminative feature learning of target samples. To tackle the aforementioned problems, we propose a novel matrix factorization-based HDA method called HDA with generalized similarity and dissimilarity regularization (HGSDR). Specifically, we propose a similarity regularizer by establishing the cross-domain Laplacian graph with label information to explore the similarity between cross-domain samples from the identical class. And we propose a dissimilarity regularizer based on the inner product strategy to expand the separability of cross-domain labeled samples from different classes. For unlabeled target samples, we keep their neighbor relationship to preserve the similarity and separability between them in the original space. Hence, the generalized similarity and dissimilarity regularization is built by integrating the above regularizers to facilitate cross-domain samples to form discriminative class distributions. HGSDR can more efficiently match the distributions of the two domains both from the global and sample viewpoints, thereby learning discriminative features for target samples. Extensive experiments on the benchmark datasets demonstrate the superiority of the proposed method against several state-of-the-art methods.
RESUMEN
Fish species identification plays a vital role in marine fisheries resource exploration, yet datasets related to marine fish resources are scarce. In open-water environments, various fish species often exhibit similar appearances and sizes. To solve these issues, we propose a few-shot learning approach to identifying fish species. Our approach involves two key components. Firstly, the embedding module was designed to address the challenges posed by a large number of fish species with similar phenotypes by utilizing the distribution relationships of species in the embedding space. Secondly, a metric function was introduced, effectively enhancing the performance of fish species classification and successfully addressing the issue of limited sample quantity. The proposed model is trained end to end on fish species public datasets including the Croatian fish dataset, Fish4Knowledge and WildFish. Compared with the prototypical networks, our method performs more effectively and improves accuracy by 2% to 10%; it is able to identify fish effectively in small samples sizes and complex scene scenarios. This method provides a valuable technological tool for the development of fisheries resources and the preservation of fish biodiversity.
RESUMEN
A surface-enhanced Raman scattering (SERS) method for measuring nitrate nitrogen in aquaculture water was developed using a substrate of ß-cyclodextrin-modified gold nanoparticles (SH-ß-CD@AuNPs). Addressing the issues of low sensitivity, narrow linear range, and relatively poor selectivity of single metal nanoparticles in the SERS detection of nitrate nitrogen, we combined metal nanoparticles with cyclodextrin supramolecular compounds to prepare a AuNPs substrate enveloped by cyclodextrin, which exhibits ultra-high selectivity and Raman activity. Subsequently, vanadium(III) chloride was used to convert nitrate ions into nitrite ions. The adsorption mechanism between the reaction product benzotriazole (BTAH) of o-phenylenediamine (OPD) and nitrite ions on the SH-ß-CD@AuNPs substrate was studied through SERS, achieving the simultaneous detection of nitrate nitrogen and nitrite nitrogen. The experimental results show that BTAH exhibits distinct SERS characteristic peaks at 1168, 1240, 1375, and 1600 cm-1, with the lowest detection limits of 3.33 × 10-2, 5.84 × 10-2, 2.40 × 10-2, and 1.05 × 10-2 µmol/L, respectively, and a linear range of 0.1-30.0 µmol/L. The proposed method provides an effective tool for the selective and accurate online detection of nitrite and nitrate nitrogen in aquaculture water.
RESUMEN
Aquatic products, as essential sources of protein, have attracted considerable concern by producers and consumers. Precise fish disease prevention and treatment may provide not only healthy fish protein but also ecological and economic benefits. However, unlike intelligent two-dimensional diagnoses of plants and crops, one of the most serious challenges confronted in intelligent aquaculture diagnosis is its three-dimensional space. Expert systems have been applied to diagnose fish diseases in recent decades, allowing for restricted diagnosis of certain aquaculture. However, this method needs aquaculture professionals and specialists. In addition, diagnosis speed and efficiency are limited. Therefore, developing a new quick, automatic, and real-time diagnosis approach is very critical. The integration of image-processing and computer vision technology intelligently allows the diagnosis of fish diseases. This study comprehensively reviews image-processing technology and image-based fish disease detection methods, and analyzes the benefits and drawbacks of each diagnostic approach in different environments. Although it is widely acknowledged that there are many approaches for disease diagnosis and pathogen identification, some improvements in detection accuracy and speed are still needed. Constructing AR 3D images of fish diseases, standard and shared datasets, deep learning, and data fusion techniques will be helpful in improving the accuracy and speed of fish disease diagnosis.
RESUMEN
Dissolved oxygen (DO) is one of the most important water quality factors. Maintaining the DO concentration at a desired level is of great value to both wastewater treatment plants (WWTPs) and aquaculture. This review covers various DO control strategies proposed by researchers around the world in the past 20 years. The review focuses on published research related to determination and control of DO concentrations in WWTPs in order to improve control accuracy, save aeration energy, improve effluent quality, and achieve nitrogen removal. The strategies used for DO control are categorized and discussed through the following classification: classical control such as proportional-integral-derivative (PID) control, advanced control such as model-based predictive control, intelligent control such as fuzzy and neural networks, and hybrid control. The review also includes the prediction and control strategies of DO concentration in aquaculture. Finally, a critical discussion on DO control is provided. Only a few advanced DO control strategies have achieved successful implementation, while PID controllers are still the most widely used and effective controllers in engineering practice. The challenges and limitations for a broader implementation of the advanced control strategies are analyzed and discussed.
Asunto(s)
Eliminación de Residuos Líquidos , Purificación del Agua , Nitrógeno/análisis , Oxígeno/análisis , Aguas ResidualesRESUMEN
Accurate identification of crop varieties is an important aspect of smart agriculture, which is not only essential for the management of later crop differences, but also has a significant effect on unmanned operations in planting scenarios such as facility greenhouses. In this study, five kinds of lettuce under the cultivation conditions of greenhouses were used as the research object, and a classification model of lettuce varieties with multiple growth stages was established. First of all, we used the-state-of-the-art method VOLO-D1 to establish a variety classification model for the 7 growth stages of the entire growth process. The results found that the performance of the lettuce variety classification model in the SP stage needs to be improved, but the classification effect of the model at other stages is close to 100%; Secondly, based on the challenges of the SP stage dataset, we combined the advantages of the target detection mechanism and the target classification mechanism, innovatively proposed a new method of variety identification for the SP stage, called YOLO-VOLO-LS. Finally, we used this method to model and analyze the classification of lettuce varieties in the SP stage. The result shows that the method can achieve excellent results of 95.961, 93.452, 96.059, 96.014, 96.039 in Val-acc, Test-acc, Recall, Precision, F1-score, respectively. Therefore, the method proposed in this study has a certain reference value for the accurate identification of varieties in the early growth stage of crops.
RESUMEN
Nitrite is common inorganic poison, which widely exists in various water bodies and seriously endangers human health. Therefore, it is very necessary to develop a fast and online method for the detection of nitrite. In this paper, we prepared an electrochemical sensor for highly sensitive and selective detection of nitrite, based on AuNPs/CS/MXene nanocomposite. The characterization of the nanocomposite was demonstrated by scanning electron microscopy (SEM), a transmission electron microscope (TEM), energy dispersive X-ray spectroscopy (EDX), X-ray diffraction (XRD), cyclic voltammetry (CV), and electrochemical impedance spectroscopy (EIS). Under the optimized conditions, the fabricated electrode showed good performance with the linear range of 0.5-335.5 µM and 335.5-3355 µM, the limit of detection is 69 nM, and the sensitivity is 517.8 and 403.2 µA mM-1 cm-2. The fabricated sensors also show good anti-interference ability, repeatability, and stability, and have the potential for application in real samples.
RESUMEN
Feature-based domain adaptation methods project samples from different domains into the same feature space and try to align the distribution of two domains to learn an effective transferable model. The vital problem is how to find a proper way to reduce the domain shift and improve the discriminability of features. To address the above issues, we propose a unified Probability-based Graph embedding Cross-domain and class Discriminative feature learning framework for unsupervised domain adaptation (PGCD). Specifically, we propose novel graph embedding structures to be the class discriminative transfer feature learning item and cross-domain alignment item, which can make the same-category samples compact in each domain, and fully align the local and global geometric structure across domains. Besides, two theoretical analyses are given to prove the interpretability of the proposed graph structures, which can further describe the relationships between samples to samples in single-domain and cross-domain transfer feature learning scenarios. Moreover, we adopt novel weight strategies via probability information to generate robust centroids in each proposed item to enhance the accuracy of transfer feature learning and reduce the error accumulation. Compared with the advanced approaches by comprehensive experiments, the promising performance on the benchmark datasets verify the effectiveness of the proposed model.
RESUMEN
The rapid and precise recognition of fish behavior is critical in perceiving health and welfare by allowing farmers to make informed management decisions on recirculating aquaculture systems while reducing labor. The conventional recognition methods are to obtain movement information by implanting sensors on the skin or in the body of the fish, which can affect the normal behavior and welfare of the fish. We present a novel nondestructive method with spatiotemporal and motion information based on deep learning for real-time recognition of fish schools' behavior. In this work, a dual-stream 3D convolutional neural network (DSC3D) was proposed for the recognition of five behavior states of fish schools, including feeding, hypoxia, hypothermia, frightening and normal behavior. This DSC3D combines spatiotemporal features and motion features by using FlowNet2 and 3D convolutional neural networks and shows significant results suitable for industrial applications in automatic monitoring of fish behavior, with an average accuracy rate of 95.79%. The model evaluation results on the test dataset further demonstrated that our proposed method could be used as an effective tool for the intelligent perception of fish health status.
RESUMEN
With the continuous expansion of aquaculture scale and density, contemporary aquaculture methods have been forced to overproduce resulting in the accelerated imbalance rate of water environment, the frequent occurrence of fish diseases, and the decline of aquatic product quality. Moreover, due to the fact that the average age profile of agricultural workers in many parts of the world are on the higher side, fishery production will face the dilemma of shortage of labor, and aquaculture methods are in urgent need of change. Modern information technology has gradually penetrated into various fields of agriculture, and the concept of intelligent fish farm has also begun to take shape. The intelligent fish farm tries to deal with the precise work of increasing oxygen, optimizing feeding, reducing disease incidences, and accurately harvesting through the idea of "replacing human with machine," so as to liberate the manpower completely and realize the green and sustainable aquaculture. This paper reviews the application of fishery intelligent equipment, IoT, edge computing, 5G, and artificial intelligence algorithms in modern aquaculture, and analyzes the existing problems and future development prospects. Meanwhile, based on different business requirements, the design frameworks for key functional modules in the construction of intelligent fish farm are proposed.
RESUMEN
Crustacean farming is a fast-growing sector and has contributed to improving incomes. Many studies have focused on how to improve crustacean production. Information about crustacean behavior is important in this respect. Manual methods of detecting crustacean behavior are usually infectible, time-consuming, and imprecise. Therefore, automatic growth situation monitoring according to changes in behavior has gained more attention, including acoustic technology, machine vision, and sensors. This article reviews the development of these automatic behavior monitoring methods over the past three decades and summarizes their domains of application, as well as their advantages and disadvantages. Furthermore, the challenges of individual sensitivity and aquaculture environment for future research on the behavior of crustaceans are also highlighted. Studies show that feeding behavior, movement rhythms, and reproduction behavior are the three most important behaviors of crustaceans, and the applications of information technology such as advanced machine vision technology have great significance to accelerate the development of new means and techniques for more effective automatic monitoring. However, the accuracy and intelligence still need to be improved to meet intensive aquaculture requirements. Our purpose is to provide researchers and practitioners with a better understanding of the state of the art of automatic monitoring of crustacean behaviors, pursuant of supporting the implementation of smart crustacean farming applications.
RESUMEN
A SPEC/AuNPs/PMB modified electrode was prepared by electrodeposition and electro-polymerization. The electrochemical behavior of reduced nicotinamide adenine dinucleotide (NADH) on the surface of the modified electrode was studied by cyclic voltammetry. A certain amount of substrate and glutamate dehydrogenase (GLDH) were coated on the modified electrode to form a functional enzyme membrane. The ammonia nitrogen in the water sample could be calculated indirectly by measuring the consumption of NADH in the reaction. The results showed that the strength of electro-catalytic current signal was increased by two times; the catalytic oxidation potential was shifted to the left by 0.5 V, and the anti-interference ability of the sensor was enhanced. The optimum substrate concentration and enzyme loading were determined as 1.3 mM NADH, 28 mM α-Ketoglutarate and 2.0 U GLDH, respectively. The homemade ceramic heating plate controlled the working electrode to work at 37 °C. A pH compensation algorithm based on piecewise linear interpolation could reduce the measurement error to less than 3.29 µM. The biosensor exhibited good linearity in the range of 0~300 µM with a detection limit of 0.65 µM NH4+. Compared with standard Nessler's method, the recoveries were 93.71~105.92%. The biosensor was found to be stable for at least 14 days when refrigerated and sealed at 4 °C.
Asunto(s)
Compuestos de Amonio , Acuicultura , Monitoreo del Ambiente/instrumentación , Contaminantes Químicos del Agua/análisis , Técnicas Biosensibles , Catálisis , Electrodos , Monitoreo del Ambiente/métodos , Oro , Nanopartículas del Metal , NAD , Oxidación-ReducciónRESUMEN
A novel "on-off" fluorescent probe was synthesized for highly sensitive and ultra-trace determination of ammonia nitrogen in aquaculture water. Ammonium can react with formaldehyde and sodium hydroxide to form a ring substance (urotropine), which shows no fluorescence signal. Palmatine hydrochloride (PAL) can enter the hydrophobic cavity of cucurbit[7]uril (CB[7]), eventually forming a 1 : 1 host guest complex called PAL@CB[7] under neutral or acidic conditions, which has strong green fluorescence with the maximum excitation (λex) wavelength at 343 nm, and the maximum emission (λem) wavelength at 500 nm, while urotropine has a fluorescence quenching effect on the fluorescence enhancement system of PAL@CB[7]. Therefore, a fluorescent chemosensor based on PAL@CB[7] and the reaction of ammonia nitrogen with formaldehyde was developed. The results indicate that the linearity range and the limit of detection of the proposed method are 1-300 µg L-1 with a good correlation coefficient (r2 = 0.9966) and 1.8 × 10-2 µg L-1, respectively. Under the optimal conditions, the method was employed for the detection of ammonia nitrogen in real aquaculture water samples, revealing high selectivity and sensitivity. In the future, the combination of the "on-off" fluorescence method, a portable hardware system and intelligent algorithms will provide technology support for the design of on-line sensors for measuring ammonia nitrogen in aquaculture water.
Asunto(s)
Amoníaco , Colorantes Fluorescentes , Acuicultura , Hidrocarburos Aromáticos con Puentes , Imidazoles , Nitrógeno , Espectrometría de Fluorescencia , AguaRESUMEN
In this paper, a new nanocomposite AuNPs/MXene/ERGO was prepared for sensitive electrochemical detection of nitrite. The nanocomposite was prepared by a facile one-step electrodeposition, HAuCl4, GO and MXene mixed in PBS solution with the applied potential of -1.4 V for 600 s. The modified material was characterized by scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDS), X-ray diffraction (XRD) and cyclic voltammetry (CV). The electrochemical behavior of nitrite at the modified electrode was performed by CV and chronoamperometry. The AuNPs/MXene/ERGO/GCE showed a well-defined oxidation peak for nitrite at +0.83 V (Vs. Ag/AgCl) in 0.1 M phosphate buffer solution (pH 7). The amperometric responses indicated the sensor had linear ranges of 0.5 to 80 µM and 80 to 780 µM with the LOD (0.15 µM and 0.015 µM) and sensitivity (340.14 and 977.89 µA mM-1 cm-2), respectively. Moreover, the fabricated sensor also showed good selectivity, repeatability, and long-term stability with satisfactory recoveries for a real sample. We also propose the work that needs to be done in the future for material improvements in the conclusion.
RESUMEN
Food scarcity, population growth, and global climate change have propelled crop yield growth driven by high-throughput phenotyping into the era of big data. However, access to large-scale phenotypic data has now become a critical barrier that phenomics urgently must overcome. Fortunately, the high-throughput plant phenotyping platform (HT3P), employing advanced sensors and data collection systems, can take full advantage of non-destructive and high-throughput methods to monitor, quantify, and evaluate specific phenotypes for large-scale agricultural experiments, and it can effectively perform phenotypic tasks that traditional phenotyping could not do. In this way, HT3Ps are novel and powerful tools, for which various commercial, customized, and even self-developed ones have been recently introduced in rising numbers. Here, we review these HT3Ps in nearly 7 years from greenhouses and growth chambers to the field, and from ground-based proximal phenotyping to aerial large-scale remote sensing. Platform configurations, novelties, operating modes, current developments, as well the strengths and weaknesses of diverse types of HT3Ps are thoroughly and clearly described. Then, miscellaneous combinations of HT3Ps for comparative validation and comprehensive analysis are systematically present, for the first time. Finally, we consider current phenotypic challenges and provide fresh perspectives on future development trends of HT3Ps. This review aims to provide ideas, thoughts, and insights for the optimal selection, exploitation, and utilization of HT3Ps, and thereby pave the way to break through current phenotyping bottlenecks in botany.
RESUMEN
Nitrite and nitrate are widely found in various water environments but the potential toxicity of nitrite and nitrate poses a great threat to human health. Recently, many methods have been developed to detect nitrate and nitrite in water. One of them is to use graphene-based materials. Graphene is a two-dimensional carbon nano-material with sp2 hybrid orbital, which has a large surface area and excellent conductivity and electron transfer ability. It is widely used for modifying electrodes for electrochemical sensors. Graphene based electrochemical sensors have the advantages of being low cost, effective and efficient for nitrite and nitrate detection. This paper reviews the application of graphene-based nanomaterials for electrochemical detection of nitrate and nitrite in water. The properties and advantages of the electrodes were modified by graphene, graphene oxide and reduced graphene oxide nanocomposite in the development of nitrite sensors are discussed in detail. Based on the review, the paper summarizes the working conditions and performance of different sensors, including working potential, pH, detection range, detection limit, sensitivity, reproducibility, repeatability and long-term stability. Furthermore, the challenges and suggestions for future research on the application of graphene-based nanocomposite electrochemical sensors for nitrite detection are also highlighted.
RESUMEN
Dissolved oxygen is an important index to evaluate water quality, and its concentration is of great significance in industrial production, environmental monitoring, aquaculture, food production, and other fields. As its change is a continuous dynamic process, the dissolved oxygen concentration needs to be accurately measured in real time. In this paper, the principles, main applications, advantages, and disadvantages of iodometric titration, electrochemical detection, and optical detection, which are commonly used dissolved oxygen detection methods, are systematically analyzed and summarized. The detection mechanisms and materials of electrochemical and optical detection methods are examined and reviewed. Because external environmental factors readily cause interferences in dissolved oxygen detection, the traditional detection methods cannot adequately meet the accuracy, real-time, stability, and other measurement requirements; thus, it is urgent to use intelligent methods to make up for these deficiencies. This paper studies the application of intelligent technology in intelligent signal transfer processing, digital signal processing, and the real-time dynamic adaptive compensation and correction of dissolved oxygen sensors. The combined application of optical detection technology, new fluorescence-sensitive materials, and intelligent technology is the focus of future research on dissolved oxygen sensors.