Drones and Deep Learning for Detecting Fish Carcasses During Fish Kills

We propose a lightweight detection algorithm based on the Single Shot MultiBox Detector (SSD) algorithm in order to facilitate sheep management and to realize sheep facial identification, and we take the self-constructed dataset as the research object. First, the SSD replaces the VGG16 backbone network with MobileNetv3, a lightweight neural network, to create a hybrid model that is much smaller. Second, the ECA attention mechanism is incorporated into the backend of the 72 × 160 bottleneck layer. Finally, the SmoothL1 loss function is substituted with the BalancedL1 loss function. The optimized model’s size decreases significantly from the original SSD’s 132 MB to just 22.4 MB. It achieves a mean average precision of 83.47% and maintains an average frame rate of 68.53 frames per second. Compared to the basic SSD model, the mean average precision has increased by 3.25 percentage points, the model size has decreased by 109.6 MB, and the detection speed has improved by 9.55 frames per second. In comparative experiments using the same dataset with different object detection models, the proposed model outperforms the SSD, Faster R-CNN, Retinanet, and CenterNet in terms of mean average precision, with improvements of 3.25 percentage points, 4.71 percentage points, 2.38 percentage points, and 8.13 percentage points, respectively. The detection speed has shown significant improvements, increasing by 9.55, 58.55, 53.1, and 12.37 frames per second, respectively. The improved model presented in this paper significantly reduces the model’s size and computational requirements while maintaining an excellent performance. This provides a valuable reference for the digitalization of animal husbandry and livestock farming.

Posture detection targeted towards providing assessments for the monitoring of health and welfare of pigs has been of great interest to researchers from different disciplines. Existing studies applying machine vision techniques are mostly based on methods using three-dimensional imaging systems, or two-dimensional systems with the limitation of monitoring under controlled conditions. Thus, the main goal of this study was to determine whether a two-dimensional imaging system, along with deep learning approaches, could be utilized to detect the standing and lying (belly and side) postures of pigs under commercial farm conditions. Three deep learning-based detector methods, including faster regions with convolutional neural network features (Faster R-CNN), single shot multibox detector (SSD) and region-based fully convolutional network (R-FCN), combined with Inception V2, Residual Network (ResNet) and Inception ResNet V2 feature extractions of RGB images were proposed. Data from different commercial farms were used for training and validation of the proposed models. The experimental results demonstrated that the R-FCN ResNet101 method was able to detect lying and standing postures with higher average precision (AP) of 0.93, 0.95 and 0.92 for standing, lying on side and lying on belly postures, respectively and mean average precision (mAP) of more than 0.93.

The ability to detect small targets and the speed of the target detector are very important for the application of remote sensing image detection, and in this paper, we propose an effective and efficient method (named CISPNet) with high detection accuracy and compact architecture. In particular, according to the characteristics of the data, we apply a context information scene perception (CISP) module to obtain the contextual information for targets of different scales and use k-means clustering to set the aspect ratios and size of the default boxes. The proposed method inherits the network structure of Single Shot MultiBox Detector (SSD) and introduces the CISP module into it. We create a dataset in the Pascal Visual Object Classes (VOC) format, annotated with the three types of detection targets, aircraft, ship, and oiltanker. Experimental results on our remote sensing image dataset as well as the Northwestern Polytechnical University very-high-resolution (NWPU VRH-10) dataset demonstrate that the proposed CISPNet performs much better than the original SSD and other detectors especially for small objects. Specifically, our network can achieve 80.34% mean average precision (mAP) at the speed of 50.7 frames per second (FPS) with the input size 300 × 300 pixels on the remote sensing image dataset. On extended experiments, the performance of CISPNet in fuzzy target detection in remote sensing image is better than that of SSD.

Purpose In order to resolve the situation of high missed diagnosis rate and high misdiagnosis rate of the pathological analysis of the gastrointestinal endoscopic images by experts, we propose an automatic polyp detection algorithm based on Single Shot Multibox Detector (SSD). Method In the paper, SSD is based on VGG-16, the fully connected layer is changed to a convolutional layer, and four convolutional layers with successively decreasing scales are added as a new network structure. In order to verify the practicability, it is not only compared with manual polyp detection but also with Mask R-CNN. Results Multiple experimental results show that the mean Average Precision (mAP) of the SSD network is 95.74%, which is 12.4% higher than the manual detection and 5.7% higher than the Mask R-CNN. When detecting a single frame of image, the detection speed of SSD is 8.41 times that of manual detection. Conclusion Based on the traditional pattern recognition algorithm and the target detection algorithm using deep learning, we select a variety of algorithms to identify and classify polyps to achieve efficient detection results. Our research demonstrates that deep learning has a lot of room for development in the field of gastrointestinal image recognition.

Forest fire risk assessment and prevention are crucial topics in environmental management. The most popular method involves using drone imagery and object detection models to analyze risk. However, traditional drone images typically use the sRGB color space, which may lose valuable information. In this study, we systematically investigate the impact of different color spaces (sRGB, Linear RGB, Log RGB, XYZ, LMS, and D-Log) on the performance of state-of-the-art vision transformer models and the latest YOLO model for tree condition detection. Our experiments demonstrate that Log RGB and Linear RGB significantly outperform the conventional sRGB color space, with Log RGB achieving a 27.16% improvement in mean average precision (mAP) and a 34.44% gain in mean average recall (mAR). These improvements are attributed to Log RGB’s enhanced dynamic range, superior illumination invariance, and better information preservation, which enable the detection of subtle environmental details crucial for early wildfire risk assessment. Overall, our findings highlight the potential of leveraging alternative color space representations to develop more accurate and robust tools for wildfire risk assessment.

Nursery plays an important role on starting chili cultivation, determining the crop health, fertility from disease, and growth performance. Early-stage germination detection is necessary to minimize nursery failure and improve plant health, but manual detection is challenging for large scale nursery in the greenhouse. The aim of this research was to develop an automatic detection model integrated with a You Only Look Once (YOLO) based deep learning algorithm using RGB camera to monitor the chili germination stages. Method to detect germination was YOLO with several steps, included: (1) early stages chili germination images acquisition, (2) datasets preparations, (3) dataset annotation and labeling, (4) model development using deep learning YOLO algorithms, and (5) model testing and validation. The training of 11,423 images was conducted utilizing the YOLOv5 and YOLOv8 algorithms, which categorized into, three classes (germinated, not germinated, and cotyledon appearance). The model was evaluated using mean Average Precision (mAP), precision, accuracy, and recall with the respective values of 0.697, 73%, 75%, and 73% for YOLOv8, and 0.664, 70%, 73%, and 70% for YOLOv5. Both model achieved high accuracy, but YOLOv8 was better to detect and classify chili seedling growth stages than YOLOv5. This study also demonstrated that model can be implemented in real applications integrated with automatic monitoring system included in the model. Keywords: Chili seedling, Deep learning, Detection system, Germination.

Optimising the cost of nutrients in fish diets is crucial for aquaculture production, as feed accounts for 50% of operational expenses in intensive aquaculture. The rising demand for mustard oil cake (MOC), a commonly used feed ingredient in Indian aquaculture, has led to a significant increase in its price due to its diverse applications. Azolla (Azolla pinnata), a free-floating freshwater plant, has attracted researchers in fish feed formulation due to its rich nutrient profile. The present study was conducted to evaluate the effects of replacement of MOC with Azolla meal in the diet of common carp (Cyprinus carpio) fry. Pelleted feed was formulated with Azolla meal replacing MOC at levels of 0, 10, 20, 30 and 40% and fed to fish. Growth performance, feed conversion ratio (FCR), haematological parameters and proximate composition in fish carcass were assessed after 90 days of experimental feeding. Simultaneously, a control feed without supplementing of Azolla meal was also prepared. Fish were fed daily with the experimental diets at 8, 6 and 4% of their body weight during the first, second and third months respectively, in the morning hours (10:00 hrs). Principal component analysis (PCA) was performed to assess the effect of formulated diets on fish carcasses composition and haematological parameters. The results showed that replacing 30% of MOC with Azolla meal in the diet, led to significantly (p≤0.05) higher specific growth rate (SGR) and percentage weight gain and significantly (p≤0.05) lower feed conversion ratio (FCR) at this concentration. Fish survival rate ranged from 80.00 to 93.33% among different treatments. Further, significant (p≤0.05) differences were observed in both haematological parameters of fish and the proximate composition of fish carcass. among the different treatments. These findings suggest that MOC can be effectively replaced with 30% Azolla meal in the diet for rearing common carp fry to fingerling stage, without compromising the growth performance. Keywords: Aquaculture, Feed conversion ratio, Health status, Mustard oil cake, Principal component analysis, Specific growth rate, Survival

The indoor scene object detection technology is of important research significance, which is one of the popular research topics in the field of scene understanding for indoor robots. In recent years, the solutions based on deep learning have achieved good results in object detection. However, there are still some problems to be further studied in indoor object detection methods, such as lighting problem and occlusion problem caused by the complexity of the indoor environment. Aiming at these problems, an improved object detection method based on deep neural networks is proposed in this article, which uses a framework similar to the single-shot multibox detector (SSD). In the proposed method, an improved ResNet50 network is used to enhance the transmission of information, and the feature expression capability of the feature extraction network is improved. At the same time, a multiscale contextual information extraction (MCIE) module is used to extract the contextual information of the indoor scene, so as to improve the indoor object detection effect. In addition, an improved dual-threshold non-maximum suppression (DT-NMS) algorithm is used to alleviate the occlusion problem in indoor scenes. Finally, the public dataset SUN2012 is further screened for the special application of indoor scene object detection, and the proposed method is tested on this dataset. The experimental results show that the mean average precision (mAP) of the proposed method can reach 54.10%, which is higher than those of the state-of-the-art methods.

Object detection with deep learning model has achieved good results in many fields, but in some fields that think highly of data privacy, such as medical care, its applications is greatly limited by data. And Federated Learning allows clients to train a model together, while leaving their data in the local, without sharing with the server or other clients. Using the methods of Federated Learning, such as Federated Averaging(FedAvg), to train models can provide privacy, security benefits. Nonetheless, there is little experiment applying Federated Learning algorithms to train the model with a large number of parameters, such as deep learning object detection model. With non-IID data, the accuracy of object detection model trained by FedAvg reduces significantly, and need more rounds to coverage. In this work, we use Kullback-Leibler divergence(KLD) measure the weights divergence between different model trained with non-IID data. And we propose a useful scheme to improve FedAvg based Abnormal Weights Supression, reducing the influence of the weights divergence caused by non-IID and unbalanced data. As a representative of object detection, we choose Single Shot MultiBox Detector(SSD) as the base model. The results of the experiments show that the Mean Average Precision(mAP) get obvious improvement in Pascal VOC 2007 test dataset.

Although transportation practitioners nowadays have an unprecedented level of access to massive information, there still exist substantial gaps in their ability to efficiently and reliably find the right information, at the right time, and for the task/decision at hand. To address this gap, this paper proposes a context-aware information retrieval (IR) approach, which can capture and exploit the conceptualization of user needs, decision context, and content meanings to support the retrieval of information that is more relevant to decision making. The proposed IR approach includes three primary components: semantic annotation (SA), semantic query processing (SQP), and semantic document ranking (SDR). This paper focuses on SA for IR for supporting the Transportation Project Environmental Review (TPER) decision-making process. The paper proposes an epistemology-based SA algorithm for automatically annotating webpages in the TPER domain with contextual concepts from an epistemological model. The TPER epistemology is a semantic model for representing and reasoning about information and information retrieval in the TPER domain. In developing the proposed algorithm, a number of syntactic-based and semantic-based annotation methods/algorithms were developed and tested. For the syntactic-based algorithms, the effect of syntactic expansion and filtering was investigated. For the semantic-based algorithms, different semantic similarity calculation methods were evaluated. All the algorithms were tested on a testing data set of 1,328 webpages, which were collected from the FHWA Environmental Review Toolkit Website, and evaluated in terms of Mean Average Precision (MAP) and average precision. The final, proposed SA algorithm achieved 84.07% MAP and 90.67% average precision at top 50 documents, on the testing data.

Timber transportation surveillance is the key and important means of forest resource management. In order to surveille timber transportation in the forest road, we carry out timber transportation vehicle detection from the video of bayonet, with which we propose an improvement of SSD (Single Shot Multibox Detector) based on MXNet. We establish a timber transportation vehicle dataset for training and testing the ResNet50 and Darknet53 with three thresholds. As mAP (mean Average Precision) is closely related to IoU (Intersection over Union), the original calculation of which does not consider overlap and disjoint between prediction box and the ground truth, we find it contributes the major part of error-detection with SSD. Inspired by this, we combine a novel metric loss calculation called GIoU (Generalized Intersection over the Union) with SSD to improve mAP, which is proved to be greatly promoted in timber transportation vehicle detection.

With the development of the new generation communication system in China, the application of intelligent transportation system is more extensive, which brings higher demands for vehicle flow detection and monitoring. Traditional traffic flow detection modes often cannot meet the high statistical accuracy requirement and high-speed detection simultaneously. Therefore, an improved Inception module is integrated into the single shot multi box detector algorithm. An intelligent vehicle flow detection model is constructed based on the improved single shot multi box detector algorithm. According to the findings, the convergence speed of the improved algorithm was the fastest. When the test sample was the entire test set, the accuracy and precision values of the improved method were 93.6% and 96.0%, respectively, which were higher than all comparison target detection algorithms. The experimental results of traffic flow statistics showed that the model had the highest statistical accuracy, which converged during the training phase. During the testing phase, except for manual statistics, all methods had the lowest statistical accuracy on motorcycles. The average accuracy and precision of the designed model for various types of images were 96.9% and 96.8%, respectively. The calculation speed of this intelligent model was not significantly improved compared to the other two intelligent models, but it was significantly higher than manual monitoring methods. Two experimental data demonstrate that the intelligent vehicle flow detection model designed in this study has higher detection accuracy. The calculation speed has no significant difference compared with the traditional method, which is helpful to the traffic flow management in intelligent transportation system.

Rapid and automated identification of blight disease in potato will help farmers to apply timely remedies to protect their produce. Manual detection of blight disease can be cumbersome and may require trained experts. To overcome these issues, we present an automated system using the Mask Region-based convolutional neural network (Mask R-CNN) architecture, with residual network as the backbone network for detecting blight disease patches on potato leaves in field conditions. The approach uses transfer learning, which can generate good results even with small datasets. The model was trained on a dataset of 1423 images of potato leaves obtained from fields in different geographical locations and at different times of the day. The images were manually annotated to create over 6200 labeled patches covering diseased and healthy portions of the leaf. The Mask R-CNN model was able to correctly differentiate between the diseased patch on the potato leaf and the similar-looking background soil patches, which can confound the outcome of binary classification. To improve the detection performance, the original RGB dataset was then converted to HSL, HSV, LAB, XYZ, and YCrCb color spaces. A separate model was created for each color space and tested on 417 field-based test images. This yielded 81.4% mean average precision on the LAB model and 56.9% mean average recall on the HSL model, slightly outperforming the original RGB color space model. Manual analysis of the detection performance indicates an overall precision of 98% on leaf images in a field environment containing complex backgrounds.

Over the last few years, researchers have focused on computer‐aided polyp detection in gastroscopy. Deep learning (DL) has shown great promise for polyps' identification. The most exceptional contribution of DL methods in gastroenterology is their ability to identify polyps quickly and accurately using convolution neural network. Nonetheless, despite significant advancements, automatic detection of small polyps remains a challenging and complex task. Furthermore, due to multiple pooling operations, the features of small polyps are lost, resulting in low detection accuracy. This paper proposes an efficient object detection method for polyp detection using gastric images to address this issue. A single‐shot multi‐box detector (SSD) was combined with the feature extractor VGG‐16, and the Refined Map Block (RMB) was integrated into SSD's high‐resolution feature maps to get more semantic information. The RMB output was used as the input to the successive layers. The RMB comprises of attention cascade and feature map concatenation cascade. The attention cascade improved the localization accuracy, while the feature map concatenation cascade improved the classification accuracy. Using the former, the proposed attention‐based SSD for gastric polyps (ASSD‐GPNet) model focused on the specific information, a polyp, rather than the background. Furthermore, the feature map concatenation cascade adds semantic information while reducing computational complexity. The output of these two cascades was combined to produce a refined feature map that enhances the detection of small polyps. The model was trained and tested on 1970 gastric images and Pascal VOC07 + 12. Image augmentation was applied to increase the training data of gastric images to reduce overfitting and skip connections were used to overcome the vanishing gradient problem. Overall, the experimental results demonstrated that the proposed model outperformed than compared models in both medical and natural images. The ASSD‐GPNet obtained mean average precision (mAP) of 94.2% on gastric images and 76.9% on Pascal VOC.

Mangrove forests provide critical ecosystem services for coastal protection, biodiversity conservation, and carbon sequestration, yet monitoring their condition at crown level remains challenging using traditional field surveys alone. This study investigates the use of high-resolution unmanned aerial vehicle (UAV) imagery and deep learning for automated delineation of mangrove crowns in Lian, Batangas, Philippines. Orthomosaic RGB images acquired at 25 ~ m altitude (GSD ${\approx} 1.34 ~ \mathrm {cm} ^ { 2}$ pixel) were tiled into $256 \times 256$ px patches and manually annotated into three structural classes: Individual Crowns (IC), Crown Clusters (CC), and Canopy Gaps (CG). To reduce systematic confusion between gaps and low-canopy vegetation, a vegetation-focused dataset variant retaining only IC and CC labels was also constructed. We fine-tuned multiple Mask R-CNN variants (R101C4, R101DC5, R101FPN, X101-32 $\times$ 8d-FPN) pretrained on COCO and benchmarked them using Average Precision (AP) across object scales. The R101FPN backbone achieved the best trade-off between accuracy and efficiency (Scenario 1), with mean AP=17.71, $\mathrm{AP}_{50}=32.43$, and shortest training time per 1k iterations. Using this backbone, a second scenario incorporating an expanded, multi-site dataset and removal of the gap class further improved overall detection performance to $\mathrm{AP}=21.71$ and $\mathrm{AP}_{50}=46.24$, at substantially reduced training time. To quantify segmentation quality beyond box-level metrics, we computed region- and boundary-aware measures for the best model: mean Dice coefficient of 0.9273, Boundary IoU of 0.5181, and symmetric Hausdorff distance of 18.54 px over 295 validation images. These results indicate accurate recovery of crown extent, with moderate boundary imprecision in dense clusters and low-contrast conditions. Overall, the findings demonstrate that UAV-derived imagery coupled with Mask R-CNN provides a practical and effective framework for operational mangrove crown mapping in the Philippines, and highlight remaining challenges related to small crowns, crown clusters, and gap representation. -Mangrove forests, UAV imagery, Deep learning, Mask R-CNN, ResNet101, Object detection, Instance segmentation