Extraction Ability Research Articles

Green fruit detection methods: Innovative application of camouflage object detection and multilevel feature mining

Accurate detection of object fruits is essential for optimizing picking efficiency and predicting fruit yields. However, detecting early-stage ripening or green fruits, especially in complex fields, poses significant challenge due to their similarity to green leaves. This study introduces the TEAVit model, a novel camouflage object detection network specifically tailored for identifying green tomatoes in intricate agricultural environments. TEAVit incorporates a texture-edge-awareness module (TEAM) to enhance the extraction ability of texture feature by combining high-level and low-level features, an edge-guided feature module (EFM) to address background complexities and occlusions, and a context-aggregation module (CAM) to leverage contextual semantics. Experimental validation results demonstrate that the S-measure, E-measure, and F-measure performance metrics all exceed 90% on the tomato dataset, accompanied by a mean absolute error of 0.0245. These findings underpinned the effectiveness of the proposed green fruit camouflage object detection algorithm in offering new insights for agricultural target localization.

Evaluation of the Antioxidant Activity and Phenolic Composition of Different Monofloral and Polyfloral Brazilian Honey Extracts.

This study aimed to evaluate the chemical composition and antioxidant activity of phenolic extracts from monofloral and polyfloral honey samples obtained from different Brazilian regions. The chemical composition (total content of phenolic compounds and flavonoids) of extracts were measured by using colorimetric assays and analyzed by high performance liquid chromatographic (HPLC-DAD). The antioxidant activity was evaluated by chemical and biochemical assays (reducing power assay, 1,1-diphenyl-2-picrylhydrazyl (DPPH⋅) and 2,2-azino-bis(3-ethylbenzothiazoline-6-sulphonic) acid (ABTS⋅+) scavenger assays. It was also investigated the ability of extracts in attenuate lipid peroxidation induced by Fe2+ in phospholipids. The obtained results clearly demonstrated that the botanical origin and geographical region of honey collection influenced the chemical composition and antioxidant activity of extracts. Furthermore, the samples were constituted by phenolic acids and flavonoids, which p-coumaric acid was predominant among the compounds identified. All samples were able to scavenge free radicals and inhibit lipid peroxidation, and good correlations were obtained between the flavonoid content and honey color. In conclusion, the obtained extracts were constituted by antioxidant compounds, which reinforce the usage of honey in human diets, and demonstrated that the region of honey collection strong influenced in the chemical composition and, consequently, its biological effect.

AgXQA: A benchmark for advanced Agricultural Extension question answering

Large language models (LLMs) have revolutionized various scientific fields in the past few years, thanks to their generative and extractive abilities. However, their applications in the Agricultural Extension (AE) domain remain sparse and limited due to the unique challenges of unstructured agricultural data. Furthermore, mainstream LLMs excel at general and open-ended tasks but struggle with domain-specific tasks. We proposed a novel QA benchmark dataset, AgXQA, for the AE domain to address these issues. We trained and evaluated our domain-specific LM, AgRoBERTa, which outperformed other mainstream encoder- and decoder- LMs, on the extractive QA downstream task by achieving an EM score of 55.15% and an F1 score of 78.89%. Besides automated metrics, we also introduced a custom human evaluation metric, AgEES, which confirmed AgRoBERTa’s performance, as demonstrated by a 94.37% agreement rate with expert assessments, compared to 92.62% for GPT 3.5. Notably, we conducted a comprehensive qualitative analysis, whose results provide further insights into the weaknesses and strengths of both domain-specific and general LMs when evaluated on in-domain NLP tasks. Thanks to this novel dataset and specialized LM, our research enhanced further development of specialized LMs for the agriculture domain as a whole and AE in particular, thus fostering sustainable agricultural practices through improved extractive question answering.

Insulator detection based on FA‐YOLO network with improved feature extraction ability

AbstractUnmanned aerial vehicle insulator detection that aims to recognize defective insulators from transmission lines has made significant progress in recent years. However, it still faces challenges, such as the complex background of aerial images and the small memory of unmanned aerial vehicles. This paper proposes a refined insulator detection algorithm that integrates the attention mechanism in YOLOv8 to improve the feature extraction ability. Specifically, this paper introduces a fast vision transformers structure in the you only look once (YOLO) v8 backbone section to enhance feature extraction by capturing local and global features. Additionally, the global attention mechanism is incorporated in the neck for additional feature extraction by merging comprehensive spatial and channel information into the output. Furthermore, we amalgamate depth‐wise convolution, graph convolution, and residual operation in the global attention mechanism module. This design can mitigate the issues of gradient vanishing or exploding and meanwhile enhance the distinction between spatial attention and channel attention. The proposed model is then applied to a public dataset and a set of real images from a specific power station, and the detection results show that it outperforms many competitors in terms of accuracy, efficiency, and memory size.

FDB-Net: Fusion double branch network combining CNN and transformer for medical image segmentation.

The rapid development of deep learning techniques has greatly improved the performance of medical image segmentation, and medical image segmentation networks based on convolutional neural networks and Transformer have been widely used in this field. However, due to the limitation of the restricted receptive field of convolutional operation and the lack of local fine information extraction ability of the self-attention mechanism in Transformer, the current neural networks with pure convolutional or Transformer structure as the backbone still perform poorly in medical image segmentation. In this paper, we propose FDB-Net (Fusion Double Branch Network, FDB-Net), a double branch medical image segmentation network combining CNN and Transformer, by using a CNN containing gnConv blocks and a Transformer containing Varied-Size Window Attention (VWA) blocks as the feature extraction backbone network, the dual-path encoder ensures that the network has a global receptive field as well as access to the target local detail features. We also propose a new feature fusion module (Deep Feature Fusion, DFF), which helps the image to simultaneously fuse features from two different structural encoders during the encoding process, ensuring the effective fusion of global and local information of the image. Our model achieves advanced results in all three typical tasks of medical image segmentation, which fully validates the effectiveness of FDB-Net.

Instance Segmentation of Lentinus edodes Images Based on YOLOv5seg-BotNet

The shape and quantity of Lentinus edodes (commonly known as shiitake) fruiting bodies significantly affect their quality and yield. Accurate and rapid segmentation of these fruiting bodies is crucial for quality grading and yield prediction. This study proposed the YOLOv5seg-BotNet, a model for the instance segmentation of Lentinus edodes, to research its application for the mushroom industry. First, the backbone network was replaced with the BoTNet, and the spatial convolutions in the local backbone network were replaced with global self-attention modules to enhance the feature extraction ability. Subsequently, the PANet was adopted to effectively manage and integrate Lentinus edodes images in complex backgrounds at various scales. Finally, the Varifocal Loss function was employed to adjust the weights of different samples, addressing the issues of missed segmentation and mis-segmentation. The enhanced model demonstrated improvements in the precision, recall, Mask_AP, F1-Score, and FPS, achieving 97.58%, 95.74%, 95.90%, 96.65%, and 32.86 frames per second, respectively. These values represented the increases of 2.37%, 4.55%, 4.56%, 3.50%, and 2.61% compared to the original model. The model achieved dual improvements in segmentation accuracy and speed, exhibiting excellent detection and segmentation performance on Lentinus edodes fruiting bodies. This study provided technical fundamentals for future application of image detection and decision-making processes to evaluate mushroom production, including quality grading and intelligent harvesting.

A Theoretical Study on the Influence of Five- and Six-Membered N-Heterocyclic Ring Side Chains of the N-Donor Extractants on Am(III)/Eu(III) Extraction and Separation.

According to the green development requirements of carbon neutrality and carbon peaking, the effective separation of lanthanides and actinides is one of the key factors for nuclear energy to become a sustainable energy source. In recent years, o-phenanthroline-based ligands have been proven to be effective in the separation of lanthanides and actinides. In this work, based on 5,9b-dihydro-4aH-cyclopenta[1,2-b:5,4-b']dipyridines, we designed six N-heterocyclic ring ligands and theoretically studied their extraction capacity and separation selectivity for the Am(III) and Eu(III) ions. Various theoretical methods were used to analyze the properties of the ligands and study the bonding nature of the ligands with the metal ions. Thermodynamic parameters were calculated to measure the extraction ability of the ligands to the metal ions and to explore the separation capacity of the ligand for the Am(III) and Eu(III) ions. The calculated results show that the five- and six-membered N-heterocyclic ring side chains of the ligands and the distribution of the N atoms on the side chain rings have obvious effects on the bonding of the ligands to metal ions and on the extraction and separation properties of the ligands for the metal ions. It was found that the extractants with six-membered ring side chains possess an extraction ability slightly better than that of the ligands with five-membered ring side chains and that the ligands with a pair of adjacent N atoms on the side chains have a stronger separation selectivity for the Am(III)/Eu(III) ions. The theoretical research in this work will help to understand the details of binding and extraction properties of similar ligands and provide insights for the future design of five- and six-membered heterocyclic ligands.

Low-light enhancement method with dual branch feature fusion and learnable regularized attention

Restricted by the lighting conditions, the images captured at night tend to suffer from color aberration, noise, and other unfavorable factors, making it difficult for subsequent vision-based applications. To solve this problem, we propose a two-stage size-controllable low-light enhancement method, named Dual Fusion Enhancement Net (DFEN). The whole algorithm is built on a double U-Net structure, implementing brightness adjustment and detail revision respectively. A dual branch feature fusion module is adopted to enhance its ability of feature extraction and aggregation. We also design a learnable regularized attention module to balance the enhancement effect on different regions. Besides, we introduce a cosine training strategy to smooth the transition of the training target from the brightness adjustment stage to the detail revision stage during the training process. The proposed DFEN is tested on several low-light datasets, and the experimental results demonstrate that the algorithm achieves superior enhancement results with the similar parameters. It is worth noting that the lightest DFEN model reaches 11 FPS for image size of 1224×1024 in an RTX 3090 GPU.Graphical

Extra Virgin Olive Oil's Main Components' Antioxidant Activity and in Silico Effect on AKT1.

The study compared the chemical composition of various olive oils from the northern Algerian province of Bejaia. The research focused on the antioxidant activities of the oil's main constituents and their ability to inhibit the AKT1 protein, which is implicated in the development of colorectal cancer. The findings revealed that all of the examined oils fell within the extra virgin olive oil (EVOO) category and exhibited a high oleic acid content, particularly for samples from wild olives. These oils include high amount of ligstroside and oleocanthal, two important phenolic compounds. Wild olive oils stand out from cultivated ones due to their higher bitterness index. In addition, these oils have the highest concentrations of α-tocopherols and the best oxidative stability. Olive oil extracts demonstrated their antioxidant properties by neutralizing DPPH and ABTS radicals and converting ferric ions (Fe3+) to ferrous ions (Fe2+) for FRAP assay. Molecular docking was applied to assess the interaction between the main compounds identified in the analyzed olive oils and the human AKT1 protein, which is involved in the genesis of colorectal cancer. The findings revealed that lutein, oleuropein aglycone, and ligstroside aglycone had the highest binding affinity for the AKT1 protein. The present study could provide the theoretical foundation for further research on the interaction between AKT1 protein and EVOO compounds.

Kidney Tumor Segmentation Based on DWR-SegFormer

Kidney cancer is a malignant tumor with a high mortality rate. The accurate segmentation of tumors from computed tomography (CT) scans can assist physicians in clinical diagnosis. We introduced a new segmentation network called DWR-SegFormer to address the challenge of accurately segmenting kidney tumors in CT images. The method involved binarizing the label maps of clear cell renal cell carcinoma and papillary renal cell carcinoma CT images for identification, and the cancer lesion area was obtained by the label so that the model could accurately identify the area and enhance the feature extraction ability. Secondly, an optimized segmentation model combining a DWR attention mechanism and SegFormer network was constructed. MiT-B0 was used as the encoder of the model to establish long-distance feature dependencies and effectively extract feature information at different resolutions. The decoder with a multi-branch DWR module was implemented to utilize multi-scale feature information effectively and enhance segmentation accuracy. Comparing the experimental results with other existing models shows that the model significantly outperformed the comparison methods in all evaluation metrics on the CT image dataset of clear cell renal cancer. Furthermore, the experimental findings highlight the robustness of the proposed model across other datasets.

TP-Transfiner: high-quality segmentation network for tea pest.

Detecting and controlling tea pests promptly are crucial for safeguarding tea production quality. Due to the insufficient feature extraction ability of traditional CNN-based methods, they face challenges such as inaccuracy and inefficiency of detecting pests in dense and mimicry scenarios. This study proposes an end-to-end tea pest detection and segmentation framework, TeaPest-Transfiner (TP-Transfiner), based on Mask Transfiner to address the challenge of detecting and segmenting pests in mimicry and dense scenarios. In order to improve the feature extraction inability and weak accuracy of traditional convolution modules, this study proposes three strategies. Firstly, a deformable attention block is integrated into the model, which consists of deformable convolution and self-attention using the key content only term. Secondly, the FPN architecture in the backbone network is improved with a more effective feature-aligned pyramid network (FaPN). Lastly, focal loss is employed to balance positive and negative samples during the training period, and parameters are adapted to the dataset distribution. Furthermore, to address the lack of tea pest images, a dataset called TeaPestDataset is constructed, which contains 1,752 images and 29 species of tea pests. Experimental results on the TeaPestDataset show that the proposed TP-Transfiner model achieves state-of-the-art performance compared with other models, attaining a detection precision (AP50) of 87.211% and segmentation performance of 87.381%. Notably, the model shows a significant improvement in segmentation average precision (mAP) by 9.4% and a reduction in model size by 30% compared to the state-of-the-art CNN-based model Mask R-CNN. Simultaneously, TP-Transfiner's lightweight module fusion maintains fast inference speeds and a compact model size, demonstrating practical potential for pest control in tea gardens, especially in dense and mimicry scenarios.

Acmella oleracea Metabolite Extraction Using Natural Deep Eutectic Solvents

For plant metabolite extraction, natural deep eutectic solvents (NADESs) have many benefits over conventional solvents and ionic liquids. These advantages include high solubility and extraction ability, a low melting point (<100 °C), low toxicity, environmental friendliness, recyclability, and better biodegradability. This study analyses a natural deep eutectic solvent for Acmella oleracea (A. oleracea) metabolite extraction, considering the following process parameters: temperature, component ratio in the eutectic solvent, water addition, solid/liquid ratio, and extraction duration. NADESs were synthesised using a simple heating method, and the synthesis of the NADESs was verified by Fourier transform infrared spectroscopy (FTIR). In terms of total polyphenol content (TPC) and flavonoid content (TFC), the betaine/propanediol ratio in a NADES of 1:3 and S/L = 1:5 yielded the highest efficiency. A value of 8.37 mg GAE/mL was obtained for TPC by ultrasound-assisted extraction with 40% water addition, 25 °C extraction temperature, and 60 min contact time. The best result in terms of TFC was 14.50 mg QE/mL obtained through ultrasound-assisted extraction with 0% water added, 25 °C extraction temperature, and 60 min contact time.

Chemical composition and antioxidant activity of Physalis angulata L. (Solanaceae) and its effect on blood clotting and biofilm formation of some wound bacterial isolates

Physalis angulata a member of the Solanaceae family, is a plant recognized for its significant nutritional and medicinal properties. Widely distributed across tropical and subtropical regions, this plant has garnered attention for its diverse biological activities. The ethanol extract from the plant is being used in this study to investigate its biological and antioxidant properties. The ability of the extract to reduce free radicals was also tested using the compound phenyl-1-picrylhydrazyl). The DPPH method was used because it is a modern color method and is considered the most common and easy to apply, in addition to its speed and superior sensitivity in testing antioxidants for samples of plant extracts, and the values were compared with the value of ascorbic acid IC50. The study also looks at the extract's ability to prevent the formation of biofilms by different pathogens, including E. coli and Staphylococcus strains that were isolated from community labs and Ramadi Teaching Hospital. The ethanol extract demonstrated a noteworthy ability to inhibit the formation of cellular membranes in staphylococcal bacteria and E. coli. The efficacy of the extract at different concentrations (150, 200, 250 mg/mL) was also assessed, revealing that a concentration of 250 mg/mL resulted in significant inhibition rates (74.3% and 61.45% for E. coli). Furthermore, the study explores the impact of ethanol extract on blood coagulation. At high concentrations (150, 200, 250 mg/mL), the extract exhibited properties akin to aspirin and warfarin, extending the clotting times in both total and partial coagulation tests (Prothrombin time, Partial thromboplastin time) at lower concentrations (100, 50, 10 mg/mL), the extract appeared to promote blood clot formation, suggesting a concentration-dependent effect on hemostasis. This dual action positions P.angulate as a potential therapeutic agent, warranting further investigation into its biomedical applications.

Ekstrak Etanol Daun Binahong (Anredera cordifolia) di Berbagai Tingkat Konsentrasi terhadap Bakteri Staphylococcus aureus dan Pseudomonas aeruginosa

Burns can cause infection, hinder healing, and increase the risk of long-term complications. Staphylococcus aureus and Pseudomonas aeruginosa often infect burn wounds in general, infections are treated with antibiotics, but continued use of antibiotics can cause bacterial resistance. Binahong leaves (Anredera cordifolia) have potential as a substitute for antibiotics because of their antibacterial properties from active compounds such as alkaloids, polyphenols, tannins, saponins and ascorbic acid. This research aims to activate the antibacterial ability of binahong leaf ethanol extract against S. aureus and P. aeruginosa. This research uses an experimental method with a control group design using diffusion and dilution methods at varying concentrations. Statistical analysis of the inhibition zone showed that there was a significance difference of <0.05 between treatment groups. In the solid diluted method, S. aureus obtained an MIC value of 10.5% and a MBC of 11.5%, whereas P. aeruginosa had an MIC of 14.5% and a MBC of 15%. The results showed that the ethanol extract of binahong leaves was more sensitive to S. aureus than P. aeruginosa.

Ultrasound-assisted sustainable extraction of bioactive phytochemicals in shallot (Allium ascalonicum L.) peel: A DoE and metabolomics combined approach

Fruits and vegetables, essential components of human nutrition, undergo considerable losses and waste during processing. Recognizing this issue, this study delves into the potential use of shallot (Allium ascalonicum L.) peels, typically discarded by-products, using ultrasound-assisted extraction (UAE), a green and effective method. Employing the Design of Experiments (DoE) methodology, the study optimized extraction parameters, obtaining an extraction time of 10 min, an amplitude of 47%, and a solid-to-liquid ratio of 1:10. The results revealed a broad diversity of potentially bioactive compounds, including flavonoids (isorhamnetin, delphinidin), phenolic acids (sinapoyltyramine), stilbenes (piceatannol 3-O-glucoside), terpenoids (gibberellin), and sulphur compounds (3-ethylthiophene). Notably, the extract exhibited strong antioxidant activity. Additionally, the study investigated the ability of shallot peel extracts to inhibit key enzymes – alpha-glucosidase (Glu), acetylcholinesterase (AChE), and tyrosinase (Tyr) – relevant in carbohydrate metabolism, memory, and skin pigmentation. While demonstrating significant Glu inhibition, no notable AChE or Tyr inhibition was observed. Correlation analysis highlighted the involvement of flavonoids and terpenoids in the bioactivities observed. This research underscores the potential of shallot by-products as a source of valuable compounds with applications in the food and pharmaceutical industries. Further exploration is encouraged to unveil specific bioactive compounds responsible for the antioxidant potential and to harness the full potential of shallot by-products in promoting health and sustainability.

Fault Diagnosis Method of Special Vehicle Bearing Based on Multi-Scale Feature Fusion and Transfer Adversarial Learning.

To address the issues of inadequate feature extraction for rolling bearings, inaccurate fault diagnosis, and overfitting in complex operating conditions, this paper proposes a rolling bearing diagnosis method based on multi-scale feature fusion and transfer adversarial learning. Firstly, a multi-scale convolutional fusion layer is designed to effectively extract fault features from the original vibration signals at multiple time scales. Through a feature encoding fusion module based on the multi-head attention mechanism, feature fusion extraction is performed, which can model long-distance contextual information and significantly improve diagnostic accuracy and anti-noise capability. Secondly, based on the domain adaptation (DA) cross-domain feature adversarial learning strategy of transfer learning methods, the extraction of optimal domain-invariant features is achieved by reducing the gap in data distribution between the target domain and the source domain, addressing the call for research on fault diagnosis across operating conditions, equipment, and virtual-real migrations. Finally, experiments were conducted to verify and optimize the effectiveness of the feature extraction and fusion network. A public bearing dataset was used as the source domain data, and special vehicle bearing data were selected as the target domain data for comparative experiments on the effect of network transfer learning. The experimental results demonstrate that the proposed method exhibits an exceptional performance in cross-domain and variable load environments. In multiple bearing cross-domain transfer learning tasks, the method achieves an average migration fault diagnosis accuracy rate of up to 98.65%. When compared with existing methods, the proposed method significantly enhances the ability of data feature extraction, thereby achieving a more robust diagnostic performance.

UTR: A UNet-like transformer for efficient unsupervised medical image registration

The existing medical image registration algorithms have the problem of low registration accuracy when processing large deformation medical images. In order to improve registration performance and utilize the global context extraction ability of Transformers without causing high computational complexity, a UNet-like Transformer model combining CNN and Transformer was constructed for 3D medical image registration tasks. We use the Efficient Global Local Attention (EGLA) mechanism to construct a Transformer encoder to further address the difficulty of modeling long-distance dependencies in existing medical image registration networks. We leverage the local modeling capabilities of CNN and the long-distance information capture capabilities of Transformer to achieve high-precision registration. The algorithm has undergone detailed validation experiments on two public datasets. The qualitative and quantitative registration results validate the effectiveness of the proposed model.

PCB Defect Detection Based on Improved Deep Learning Model

Printed circuit boards (PCBs) play a critical role in electronic products. Ensuring these products’ long-term reliability and consistent performance requires effective PCB defect detection. Although existing deep learning models for PCB defect detection are not highly accurate, they often neglect capability considerations. This paper introduces a precise, fast, and lightweight defect detection model, CCG-YOLO, based on an enhanced YOLOv5 model to address this issue. The enhancements in CCG-YOLO can be summarized as follows: (1) Improved Backbone network: The feature extraction ability of the Backbone network is enhanced by introducing a C3HB module, which fosters spatial interaction capabilities. (2) Lightweight feature fusion network: A lightweight convolution structure called Ghost-Shuffle Convolution is incorporated in the feature fusion network, remarkably reducing model parameters while maintaining performance. (3) Efficient residual networking: To enhance model performance further, a CNeB module is introduced based on the ConvNeXt network, which replaces the C3 module in the Neck. CNeB improves model detection accuracy and reduces the number of model parameters. The combination of these enhancements results in impressive performance. CCG-YOLO achieves mean average precision (mAP@0.5) of 99.5% and 88.75% in mAP@0.5:0.95 on the TDD-Net public dataset. Compared with the original YOLOv5s algorithm, CCG-YOLO offers a 4.24% improvement in mAP@0.5:0.95, a 1[Formula: see text]MB reduction in model size, a 0.472[Formula: see text]M decrease in the number of parameters, a 0.6G floating point operation reduction in computational complexity, and a 120 frames per second real-time inference speed. These experimental results underscore that the proposed model excels in accuracy and speed and has a compact size for PCB defect detection. Moreover, CCG-YOLO is easily deployable on low-end devices, making it well-suited for meeting the real-time requirements of industrial defect detection.

DGA-grafting pyridine for ultra-selective and prior extraction of 99TcO4− from simulated spent nuclear fuel

Selective and prior extraction of 99TcO4− ahead of uranium and plutonium separation is a beneficial strategy for the modern nuclear fuel cycle. Herein, a novel DGA-grafting pyridine ligand BisDODGA-DAPy (L1) was tailored for the efficient separation of TcO4− from simulated spent nuclear fuel based on the selectivity of pyridine and synergistic effect of diglycolamide (DGA) group. Compared to the ligands BisDOSCA-DAPy (L2) and BisDODGA-MPDA (L3) with similar structure, BisDODGA-DAPy (L1) demonstrated the better separation performance including good extraction efficiency, reusability, and high loading capacity for TcO4− under high acidic medium. The interactions of the ligands with Tc(VII)/Re(VII) have been investigated in detail using FT-IR, 1H NMR titration, UV-Vis spectrophotometric titration, ESI-HRMS and DFT simulations. The extraction mechanism affected by the protonation of ligand was elucidated under different acidity. BisDODGA-DAPy (L1) demonstrated the ultra-selective extraction ability for TcO4− from simulated spent nuclear fuel. The maximum SFTc/U and SFTc/Pu values were up to 1.29 × 104 and 5.08 × 103, respectively. In the presence of 9 × 104-fold excess of NO3−, the extraction of TcO4− was almost unaffected. Moreover, the good radiolytic stability further highlights the promising potential of this ligand for 99Tc separation. DFT calculation revealed the dominant role of DAPy and DODGA in TcO4− extraction, providing the theoretical evidence for BisDODGA-DAPy (L1) to selectively bind TcO4− over NO3−.

Dense object detection methods in RAW UAV imagery based on YOLOv8

Accurate, fast and lightweight dense target detection methods are highly important for precision agriculture. To detect dense apricot flowers using drones, we propose an improved dense target detection method based on YOLOv8, named D-YOLOv8. First, we introduce the Dense Feature Pyramid Networks (D-FPN) to enhance the model’s ability to extract dense features and Dense Attention Layer (DAL) to focus on dense target areas, which enhances the feature extraction ability of dense areas, suppresses features in irrelevant areas, and improves dense target detection accuracy. Finally, RAW data are used to enhance the dataset, which introduces additional original data into RAW images, further enriching the feature input of dense objects. We perform validation on the CARPK challenge dataset and constructed a dataset. The experimental results show that our proposed D-YOLOv8m achieved 98.37% AP, while the model parameters were only 13.2 million. The improved network can effectively support any task of dense target detection.