Classification Of Species Research Articles

Automatic Wood Species Classification and Pith Detection in Log CT Images

This article focuses on the need for digitalization in the forestry and timber sector using information from CT scans of logs. The National Forest Centre (Slovak Republic) operates a unique 3D CT scanner for wooden logs at the Stráž Biotechnology Park. This real-time scanner generates a 3D model of a log, displaying the wood’s internal features/defects. To optimize log-cutting plans effectively, it is necessary to automatically detect and classify these features and defects in real time, leveraging computer vision principles. Artificial intelligence, specifically neural networks, addresses this need by enabling solutions for tasks of this nature. Building a highly efficient neural network for detecting wood features and defects requires creating a database of log scans and training the network on these data. This is a time-intensive process, as it involves manually marking internal features and defects on hundreds of CT scans of various wood types. A functional neural network for detecting internal wood defects represents a significant advancement in sector digitalization, paving the way for further automation and robotization in wood processing. For the forestry sector to remain competitive, efficiently process raw materials, and improve product quality, the effective application of CT scanning technology is essential. This technological innovation aligns the sector more closely with leaders in other fields, such as the automotive, engineering, and metalworking industries.

Dominant Tree Species Mapping Using Machine Learning Based on Multi-Temporal and Multi-Source Data

Timely and accurate tree species mapping is crucial for forest resource inventory, supporting management, conservation biology, and ecological restoration. This study utilized Sentinel-1 and Sentinel-2 data to classify five dominant tree species in Chengde and Beijing. To effectively capture the influence of multi-temporal data, data were acquired in March, June, September, and December 2020, extracting various features, including bands, spectral indices, texture features, and topographic variables. The optimal input variable combination was explored using 1519 field survey samples for training and testing datasets. Classification employed Random Forest, XGBoost, and deep learning models, with performance evaluated through out-of-bag estimation and cross-validation. The XGBoost model achieved the highest accuracy of 81.25% (kappa = 0.74) when using Sentinel-1 and Sentinel-2 bands, indices, texture features, and DEM data. Results demonstrate the effectiveness of using Sentinel data for tree species classification and emphasize the value of machine learning algorithms. This study underscores the potential of combining synthetic aperture radar (SAR) and optical data for large-scale tree species classification, with significant implications for forest monitoring and management.

Tree Species Classification for Shelterbelt Forest Based on Multi-Source Remote Sensing Data Fusion from Unmanned Aerial Vehicles

Tree Species Classification for Shelterbelt Forest Based on Multi-Source Remote Sensing Data Fusion from Unmanned Aerial Vehicles

Integrating Drone-Based LiDAR and Multispectral Data for Tree Monitoring

Forests are critical for providing ecosystem services and contributing to human well-being, but their health and extent are threatened by climate change, requiring effective monitoring systems. Traditional field-based methods are often labour-intensive, costly, and logistically challenging, limiting their use for large-scale applications. Drones offer advantages such as low operating costs, versatility, and rapid data collection. However, challenges remain in optimising data processing and methods to effectively integrate the acquired data for forest monitoring. This study addresses this challenge by integrating drone-based LiDAR and multispectral data for forest species classification and health monitoring. We developed the methodology in Ticino Park (Italy), where intensive field campaigns were conducted in 2022 to collect tree species compositions, the leaf area index (LAI), canopy chlorophyll content (CCC), and drone data. Individual trees were first extracted from LiDAR data and classified using spectral and textural features derived from the multispectral data, achieving an accuracy of 84%. Key forest traits were then retrieved from the multispectral data using machine learning regression algorithms, which showed satisfactory performance in estimating the LAI (R2 = 0.83, RMSE = 0.44 m2 m−2) and CCC (R2 = 0.80, RMSE = 0.33 g m−2). The retrieved traits were used to track species-specific changes related to drought. The results obtained highlight the potential of integrating drone-based LiDAR and multispectral data for cost-effective and accurate forest health monitoring and change detection.

3D-CNN with Multi-Scale Fusion for Tree Crown Segmentation and Species Classification

Natural secondary forests play a crucial role in global ecological security, climate change mitigation, and biodiversity conservation. However, accurately delineating individual tree crowns and identifying tree species in dense natural secondary forests remains a challenge. This study combines deep learning with traditional image segmentation methods to improve individual tree crown detection and species classification. The approach utilizes hyperspectral, unmanned aerial vehicle laser scanning data, and ground survey data from Maoershan Forest Farm in Heilongjiang Province, China. The study consists of two main processes: (1) combining semantic segmentation algorithms (U-Net and Deeplab V3 Plus) with watershed transform (WTS) for tree crown detection (U-WTS and D-WTS algorithms); (2) resampling the original images to different pixel densities (16 × 16, 32 × 32, and 64 × 64 pixels) and inputting them into five 3D-CNN models (ResNet10, ResNet18, ResNet34, ResNet50, VGG16). For tree species classification, the MSFB combined with the CNN models were used. The results show that the U-WTS algorithm achieved a recall of 0.809, precision of 0.885, and an F-score of 0.845. ResNet18 with a pixel density of 64 × 64 pixels achieved the highest overall accuracy (OA) of 0.916, an improvement of 0.049 over the original images. After incorporating MSFB, the OA improved by approximately 0.04 across all models, with only a 6% increase in model parameters. Notably, the floating-point operations (FLOPs) of ResNet18 + MSFB were only one-eighth of those of ResNet18 with 64 × 64 pixels, while achieving similar accuracy (OA: 0.912 vs. 0.916). This framework offers a scalable solution for large-scale tree species distribution mapping and forest resource inventories.

Determination and Comparison of Volatile Organic Compounds (VOCs) in Celery (Apium graveolens L.) Seeds by Headspace-Solid-Phase Microextraction (HS-SPME) Gas Chromatography-Mass Spectrometry (GC-MS) and Multivariate Analysis

The utilization of celery seeds as seasonings and in herbal teas is widespread due to their significant nutritional and medicinal value. The determination of volatile organic compounds (VOCs) is a valuable tool for species identification and classification. Currently, there is an absence of information investigating the volatile constituents present in the cultivars of celery seeds. To analyze the varietal differences of celery seeds, VOCs were isolated using headspace solid-phase microextraction with by determination by gas chromatography-mass spectrometry. The samples underwent further comparison using multivariate statistical analysis to determine the changes in VOCs. Cluster analysis and principal component analysis were used to conduct correlation measurements on the samples. An orthogonal partial least squares discriminant analysis (OPLS-DA) model was developed to discriminate between samples and identify differential compounds. The distribution of these key compounds in these seeds was analyzed using a Sankey diagram. The samples contained 30 volatile components, mainly terpenes and esters. Correlation analysis revealed distinct variations in the VOCs. Furthermore, the utilization of the OPLS-DA model facilitated the identification of 11 differential compounds. The Sankey diagram illustrated significant disparities in the distribution patterns of these key compounds in the celery seeds.

Characterizing the spectral-temporal signatures of eastern Hemlock (Tsuga canadensis) using sentinel-2 satellite images and phenology modelling

Hemlock woolly adelgid (HWA) is an invasive insect that affects the eastern hemlock population in North America, causing severe die-off and altering ecosystem dynamics. Understanding the distribution of eastern hemlock will improve future HWA management and protection of existing eastern hemlock populations. To determine the degree to which different forest types and species can be distinguished at the stand level with variable densities of eastern hemlock present, a Bayesian phenology model was used to compute seven phenological parameters from four spectral indices derived from Sentinel-2 time series imagery. We tested spectral and phenological parameters derived using this method across three classification levels, including broad forest type, hemlock density, and dominant or co-dominant evergreen species. Using Kruskal-Wallis with post-hoc Dunn’s test, we found that phenological parameters derived from the Inverted Red-Edge Chlorophyll Index and the Soil-Adjusted Vegetation Index provided the highest separability between groups across all three levels of classification. The seasonal minimum greenness and fall inflection day provided the highest degree of separability among hemlock density classes. Seasonal minimum greenness provided the highest degree of separability among evergreen species. Among the nine evergreen dominant or co-dominant species classes tested, hemlock stands were found to be separable from four of the classes. White pine stands and black spruce stands showed the highest degree of overall separability. This study demonstrates the potential for phenological parameters in stand-level evergreen species classification. The combination of Sentinel-2 time series and phenological modeling has the potential to enhance tree species mapping studies at regional scales.

Relationship between the octane number and the chemical composition of gasoline: study by FTIR spectroscopy

The oil refining process undergoes various heat treatments, during which several cuts are obtained, including cuts of gasoline of different chemical composition. Determination of physicochemical properties is essential for our fuel classification. In the present study and by exploiting a rarely used spectroscopic method in the field of production and characterization of fuels, FTIR Fourier transform infrared spectroscopy (a method characterized by its reliability and speed), we identified a correlation between the molecular composition of the fuel (translated by chemical number) and its octane number NO (a quite important parameter for species classification). The results of AR1 aromaticity index show a strong linear relationship between the latter and its NO octane number. In addition, NO also occurs in many modes depending on two indicators DOC (degree of condensation) and AR1. Indeed, the ACL aliphatic chain length index varies depending on the quality of the fuel and precisely on the subsequent fuel production process. Furthermore, this method illustrates the molecular composition of the fuel, the correlation determined between the chemical indices and the octane number offering the possibility of predicting this physico-chemical parameter.

A deep learning model for detecting and classifying multiple marine mammal species from passive acoustic data

Underwater passive acoustics is used worldwide for multi-year monitoring of marine mammals. Yet, the large amount of audio recordings raises the need to automate the detection of acoustic events. For instance, the increasing number of Offshore Wind Farms (OWF) raises key environmental and societal issues relating to their impacts on wildlife. In this context, monitoring marine mammals along with information on their acoustic environment throughout the OWF life cycle is crucial. The objective of this study is to evaluate the ability of a single deep learning model to precisely detect and localize, in time and in frequency, the marine mammal sounds over a wide frequency range and classify them by species and sound types.A broadband hydrophone, deployed at the Fécamp OWF (Normandy, France), recorded the underwater soundscape including sounds from marine mammals occurring in the area. To visualize these sounds, 15-s spectrograms were computed. From these images, dolphin (D) and porpoise (P) sounds were manually annotated, including different types of sounds: Click-Trains (DCT, PCT), Buzzes (DB, PB) and Whistles (DW). The spectrograms were then split into five-fold cross-validation datasets, each containing one half of manual annotations and one half of only background noise. A Faster R-CNN model was trained to precisely detect and classify the marine mammal sounds in the spectrograms.Three model output configurations were used: (1) overall detection of marine mammals (presence vs. absence), (2) detection and classification of species (two classes: dolphin, porpoise) and (3) sound types (five classes: DCT, DB, DW, PCT, PB). For the simplest configuration (1) 15.4 % of the spectrogram dataset had detections while missing only 6.6 % of annotated spectrograms. For the more precise configurations, (2) and (3), the mean Average Precision (mAP) achieved were 92.3 % (2) and 84.3 % (3), and the macro average Area under the curve (AUC) 95.7 % (2) and 94.9 % (3).This model will help to speed up the annotation processes, by reducing the spectrogram quantity to be manually analyzed and having time-frequency boxes already drawn. Several model parameters can be adjusted to trade off missed detections and false positives which need to be carefully considered and adapted to the problem. For instance, these adjustments would be particularly relevant depending on the human resources available to manually check the model detections and the criticality of missing marine mammal sounds. These models are promising, ranging from the simple detection of marine mammal presence to precise ecological inferences over the long term.

A New Bivalve Species Glauconome huangheensis of the Genus Glauconome J. E. Gray, 1828 (Bivalvia, Venerida, Cyrenoidea, Glauconomidae), from Shandong, China.

The family Glauconomidae has few species, limited molecular data description, and insufficient research attention. The biodiversity of Glauconomidae within China deserves further exploration. In recent years, the taxonomic status of Glauconomidae has undergone changes, and some studies have found a close relationship between Glauconomidae and the family Cyrenidae based on molecular data, suggesting that Glauconomidae should be classified under the superfamily Cyrenoidea. However, both domestic and international research has mainly focused on only four species of Glauconomidae, indicating an urgent need for more species data support. Recently, 46 specimens of Glauconomidae were collected in the Yellow River Estuary in Dongying City of Shandong Province in China. Through a comparative analysis of shell morphology and molecular phylogenetic analysis of COI and 16S rRNA, two species of Glauconomidae was discovered. One is Glauconome angulata Reeve, 1844, and the other is a new species of Glauconomidae found in the Yellow River, named Glauconome huangheensis sp. nov. The G. huangheensis sp. nov. exhibits distinct differences in shell shape and shell color compared to other species of Glauconomidae, resembling G. angulata. There are also significant differences in shell color, shell sculpture, ligament size, and shell thickness. Furthermore, the molecular phylogenetic analysis based on COI and 16S rRNA genes supports the validity of G. huangheensis sp. nov. as a species. It indicates a close phylogenetic relationship with G. angulata, making them sister species. This study provides a redescription of the morphological characteristics of G. angulata and G. huangheensis sp. nov., laying the foundation for the morphological classification, biodiversity research, and conservation of Glauconomidae species.

Transfer Learning for Wildlife Classification: Evaluating YOLOv8 against DenseNet, ResNet, and VGGNet on a Custom Dataset

This study evaluates the performance of various deep learning models, specifically DenseNet, ResNet, VGGNet, and YOLOv8, for wildlife species classification on a custom dataset. The dataset comprises 575 images of 23 endangered species sourced from reputable online repositories. The study utilizes transfer learning to fine-tune pre-trained models on the dataset, focusing on reducing training time and enhancing classification accuracy. The results demonstrate that YOLOv8 outperforms other models, achieving a training accuracy of 97.39% and a validation F1-score of 96.50‘%. These findings suggest that YOLOv8, with its advanced architecture and efficient feature extraction capabilities, holds great promise for automating wildlife monitoring and conservation efforts.

Exploring Fish Species Exploring Fish Species Classification using Deep Learning

Fishes are intriguing creatures that live in a wide range of aquatic settings, including freshwater rivers and lakes, oceans, and deep-sea trenches. Freshwater fish are considered a poor man's protein supplement since they are readily available in lakes, rivers, natural ponds, rice fields, beels, and fisheries. Many freshwater fish species resemble one another, making it difficult to classify them based on their exterior appearance. Manual fish species identification is always error-prone since it requires knowledge. Recently, computer vision and deep learning have played a key role in underwater species classification and detection studies. This research focuses on fish species categorization with Convolutional Neural Networks (CNNs) and two popular architectures, ResNet50 and VGG16. The goal of this research is to create an automated system that can correctly recognize and classify different fish species based on dataset. The two architecture ResNet50 and VGG16 are fine-tuned and CNN for training and validation of the collected fish data. The suggested method entails preprocessing the image data, training the CNN models with ResNet50 and VGG16 architectures, and assessing their performance using measures such as accuracy, precision, recall, and F1 score. This study reports the best overall classification accuracy, precision, and recall rate, which is almost 95%. Comprehensive empirical investigation has proven that when the weight and bias learning rate increases, so does the classifier's validation loss.

Artificial Intelligence-Driven Smart Aquaculture: Revolutionizing Sustainability through Automation and Machine Learning

AI incorporation in aquaculture has transformed the industry completely, making crucial processes automated, maximizing productivity, and promoting sustainability. AI, specifically machine learning, refers to the application of modern smart aquaculture systems for tasks such as fish species classification, health monitoring, feed regulation, and management of water quality. It thereby sets inefficiency issues right while reducing impacts on the environment through real-time data-driven decision-making. This article deals with very recent developments in the applications of AI and machine learning in aquaculture, pointing out their importance in increasing production as well as eco-friendly management of aquatic environments

Classification and microbes involved in Plastic biodegradation: A review

Plastic is composed of synthetic and semi-synthetic compounds with immense applications. While Plastics are dangerous and harm the environment, their practice is causing their ingesting to grow. Outdated methods of plastic degradation through physical and chemical means like landfill, incineration, and recycling are expensive and perilous and release hazardous chemical substances to the environment. Thus, proper approaches are required to overcome this trick. Biodegradation through microbes on the other hand offers a hopeful solution. Presently, numerous categories of bacteria and other microbial species degrade plastic materials. Several steps are required in plastic biodegradation like biodeterioration, depolymerization, assimilation and mineralization. These processes will result in secretion of enzymes that degrade plastics and converts it into gasses and biomass along other biproducts.

Classification of Tree Species Using Point Cloud Features from Terrestrial Laser Scanning

Classification of Tree Species Using Point Cloud Features from Terrestrial Laser Scanning

Wood species classification in open set using an improved NNO classifier

A wood species classification scheme was developed based on open set using an improved Nearest Non-Outlier (NNO) classifier. Near infrared (NIR) spectral curves were collected in spectral band 950 to 1650 nm by a micro spectrometer. The spectral dimension reduction was performed with a Metric Learning (ML) algorithm. Two improvements were proposed in the following NNO classifier. First, a cluster analysis was performed in each wood class by using a Density Peak Clustering (DPC) algorithm to get 1 to 3 clusters. A fixed threshold for all wood classes was replaced by a variable for all clusters. This threshold defines an internal boundary for one wood species to further compute a class membership score for all wood species. The classification accuracy based on these clusters of each wood class was better than that based on each class. The experimental results in different open set scenarios demonstrate that the improved NNO classifier outperformed the original NNO classifier and some other state-of-the-art open set recognition (OSR) algorithms.

Detection and Characterization of Visceral Anisakid Nematodes in Blue Whiting from Portuguese Waters.

This study employs molecular detection techniques, including conventional PCR and Sanger sequencing, to investigate the prevalence, species composition and public health implications of Anisakid nematodes in blue whiting (Micromesistius poutassou) caught off the Portuguese coast. With Portugal's high fish consumption rates and increasing preference for raw or undercooked seafood, the risk of parasitic infections, particularly anisakidosis, is rising. Fifty blue whiting fish were examined, showing a 100% infection rate with Anisakid larvae. Molecular analysis identified 68.1% of the larvae as Anisakis simplex, 18.1% as Anisakis pegreffii, and 13.8% as Hysterothylacium aduncum, marking the first report of H. aduncum in blue whiting in Portugal. Phylogenetic analysis based on the internal transcribed spacer (ITS) 1, 5.8S ribosomal RNA and ITS-2 confirmed the species classification. Notably, 42.9% of the fish were infected with multiple Anisakid species, increasing the risk of allergenic sensitization. Statistical analysis showed no significant correlation between fish width and parasitic load, and a weak negative correlation was found between fish length and parasitic load. The study contributes to food safety by integrating molecular tools that enable rapid and accurate species identification, offering new insights into the detection of biological contaminants in seafood. These findings are significant considering the rising trend in raw seafood consumption, underscoring the urgent need for enhanced detection strategies and broader parasite monitoring programs to mitigate public health risks. The high prevalence of parasitized fish highlights the necessity for the implementation of safe cooking practices to reduce the risk of anisakidosis. Further research into the allergenic potential of Hysterothylacium spp. and the ecological factors influencing this nematode distribution is recommended.

TREE DIVERSITY AND RESILIENCE IN BUCHAREST URBAN PARKS

Urban parks are essential for maintaining biodiversity in cities. This study assesses the tree diversity and ecological resilience, and stress tolerance of species in five major parks in Bucharest: Herăstrău, Kiseleff, Tineretului, Cişmigiu, and Carol. A total of 125 species from 32 families were recorded, with the highest diversity in Cişmigiu and Carol and the lowest in Kiseleff. NMDS analysis revealed similar species compositions in all parks except Kiseleff, which was distinct. No significant correlations were found between species richness and park size or age, suggesting other factors, such as management, play a larger role. Species with higher tolerance to drought and air pollution are significantly more present in Bucharest’s parks, while pest tolerance shows no significant difference in their presence. The predominance of non-native species, particularly invasive ones in Cişmigiu and Tineretului, underscores the need for better management. ANOVA results confirmed significant differences in the distribution of non-native species categories, emphasizing the importance of targeted strategies to control invasive species and promote native biodiversity.

A Neuro-Symbolic Framework for Tree Crown Delineation and Tree Species Classification

Neuro-symbolic models combine deep learning and symbolic reasoning to produce better-performing hybrids. Not only do neuro-symbolic models perform better, but they also deal better with data scarcity, enable the direct incorporation of high-level domain knowledge, and are more explainable. However, these benefits come at the cost of increased complexity, which may deter the uninitiated from using these models. In this work, we present a framework to simplify the creation of neuro-symbolic models for tree crown delineation and tree species classification via the use of object-oriented programming and hyperparameter tuning algorithms. We show that models created using our framework outperform their non-neuro-symbolic counterparts by as much as two F1 points for crown delineation and three F1 points for species classification. Furthermore, our use of hyperparameter tuning algorithms allows users to experiment with multiple formulations of domain knowledge without the burden of manual tuning.

AI-Powered Biodiversity Assessment: Species Classification via DNA Barcoding and Deep Learning

Only 1.2 million out of an estimated 8.7 million species on Earth have been fully classified through taxonomy. As biodiversity loss accelerates, ecologists are urgently revising conservation strategies, but the “taxonomic impediment” remains a significant barrier, limiting effective access to and understanding of taxonomic data for many researchers. As sequencing technologies advance, short DNA sequence fragments increasingly serve as DNA barcodes for species identification. Rapid acquisition of DNA sequences from diverse organisms is now possible, highlighting the increasing significance of DNA sequence analysis tools in species identification. This study introduces a new approach for species classification with DNA barcodes based on an ensemble of deep neural networks (DNNs). Several techniques are proposed and empirically evaluated for converting raw DNA sequence data into images fed into the DNNs. The best-performing approach is obtained by representing each pair of DNA bases with the value of a related physicochemical property. By utilizing different physicochemical properties, we can create an ensemble of networks. Our proposed ensemble obtains state-of-the-art performance on both simulated and real datasets.