Faster Region Convolutional Neural Network Research Articles

An object detection-based model for automated screening of stem-cells senescence during drug screening

Deep learning-based cell senescence detection is crucial for accurate quantitative analysis of senescence assessment. However, senescent cells are small in size and have little differences in appearance and shape in different states, which leads to insensitivity problems such as missed and false detection. In addition, complex intelligent models are not conducive to clinical application. Therefore, to solve the above problems, we proposed a Faster Region Convolutional Neural Network (Faster R-CNN) detection model with Swin Transformer (Swin-T) and group normalization (GN), called STGF R-CNN, for the detection of different senescent cells to achieve quantification assessment of induced pluripotent stem cell-derived mesenchymal stem cells (iP-MSCs) senescence. Specifically, to enhance the representation learning ability of the network, Swin-T with a hierarchical structure was constructed. It utilizes a local window attention mechanism to capture features of different scales and levels. In addition, the GN strategy is adopted to achieve a lightweight model. To verify the effectiveness of the STGF R-CNN, a cell senescence dataset, the iP-MSCs dataset, was constructed, and a series of experiments were conducted. Experiment results show that it has the advantage of high senescent detection accuracy, mean Average Precision (mAP) is 0.835, Params is 46.06M, and FLOPs is 95.62G, which significantly reduces senescent assessment time from 12 h to less than 1 s. The STGF R-CNN has advantages over existing cell senescence detection methods, providing potential for anti-senescent drug screening. Our code is available at https://github.com/RY-97/STGF-R-CNN.

D2-SPDM: Faster R-CNN-Based Defect Detection and Surface Pixel Defect Mapping with Label Enhancement in Steel Manufacturing Processes

The steel manufacturing process is inherently continuous, meaning that if defects are not effectively detected in the initial stages, they may propagate through subsequent stages, resulting in high costs for corrections in the final product. Therefore, detecting surface defects and obtaining segmentation information is critical in the steel manufacturing industry to ensure product quality and enhance production efficiency. Specifically, segmentation information is essential for accurately understanding the shape and extent of defects, providing the necessary details for subsequent processes to address these defects. However, the time-consuming and costly process of generating segmentation annotations poses a significant barrier to practical industrial applications. This paper proposes a cost-efficient segmentation labeling framework that combines deep learning-based anomaly detection and label enhancement to address these challenges in the steel manufacturing process. Using ResNet-50, defects are classified, and faster region convolutional neural networks (faster R-CNNs) are employed to identify defect types and generate bounding boxes indicating the defect locations. Subsequently, recursive learning is performed using the GrabCut algorithm and the DeepLabv3+ model based on the generated bounding boxes, significantly reducing annotation costs by generating segmentation labels. The proposed framework effectively detects defects and accurately defines them, even in randomly collected images from the steel manufacturing process, contributing to both quality control and cost reduction. This study presents a novel approach for improving the quality of the steel manufacturing process and is expected to enhance overall efficiency in the steel manufacturing industry.

AI based early identification and severity detection of nutrient deficiencies in coconut trees

Coconut trees are vital in various agricultural and economic sectors. Their susceptibility to nutrient deficiencies poses a significant threat to growth and productivity. Traditional nutrient assessment methods are time and labour-intensive, relying on manual inspection and chemical testing. There has been no significant research on detecting nutrient deficiencies in coconut trees. Additionally, assessing deficiency, and severity automatically and recommending suitable fertilizers remain unexplored. This research leverages the YOLOv9 model to identify macro and micronutrient deficiencies in coconut trees and proposes an Image Analysis based Severity Detection (IASD), to determine the severity of these deficiencies. Along with these a Severity Index Calculation Model (SICM) is also introduced that calculates the Severity Index (SI) of these deficiencies. For each identified deficiency, the appropriate fertilizer and its application quantity are suggested. Four deep learning models—RetinaNet, Faster Regional Convolutional Neural Network (Faster R-CNN), You Only Look Once version 5 (YOLOv5), and version 9 (YOLOv9) —were compared for the prediction of nutrient deficiencies in coconut trees using a dataset of 5,720 images of nutrient-deficient coconut tree leaves. YOLOv9 outperformed other models with Accuracy, Precision, and Recall values of 80 %, 98.59 %, and 80.37 %, respectively. Manual verification ensured the correctness of IASD and SICM predictions during model creation, providing farmers and agricultural professionals with a precise, automated tool for managing coconut plantations.

Automated diagnosis of atherosclerosis using multi-layer ensemble models and bio-inspired optimization in intravascular ultrasound imaging.

Atherosclerosis causes heart disease by forming plaques in arterial walls. IVUS imaging provides a high-resolution cross-sectional view of coronary arteries and plaque morphology. Healthcare professionals diagnose and quantify atherosclerosis physically or using VH-IVUS software. Since manual or VH-IVUS software-based diagnosis is time-consuming, automated plaque characterization tools are essential for accurate atherosclerosis detection and classification. Recently, deep learning (DL) and computer vision (CV) approaches are promising tools for automatically classifying plaques on IVUS images. With this motivation, this manuscript proposes an automated atherosclerotic plaque classification method using a hybrid Ant Lion Optimizer with Deep Learning (AAPC-HALODL) technique on IVUS images. The AAPC-HALODL technique uses the faster regional convolutional neural network (Faster RCNN)-based segmentation approach to identify diseased regions in the IVUS images. Next, the ShuffleNet-v2 model generates a useful set of feature vectors from the segmented IVUS images, and its hyperparameters can be optimally selected by using the HALO technique. Finally, an average ensemble classification process comprising a stacked autoencoder (SAE) and deep extreme learning machine (DELM) model can be utilized. The MICCAI Challenge 2011 dataset was used for AAPC-HALODL simulation analysis. A detailed comparative study showed that the AAPC-HALODL approach outperformed other DL models with a maximum accuracy of 98.33%, precision of 97.87%, sensitivity of 98.33%, and F score of 98.10%.

A Multi-Scale Target Detection Method Using an Improved Faster Region Convolutional Neural Network Based on Enhanced Backbone and Optimized Mechanisms.

Currently, existing deep learning methods exhibit many limitations in multi-target detection, such as low accuracy and high rates of false detection and missed detections. This paper proposes an improved Faster R-CNN algorithm, aiming to enhance the algorithm's capability in detecting multi-scale targets. This algorithm has three improvements based on Faster R-CNN. Firstly, the new algorithm uses the ResNet101 network for feature extraction of the detection image, which achieves stronger feature extraction capabilities. Secondly, the new algorithm integrates Online Hard Example Mining (OHEM), Soft non-maximum suppression (Soft-NMS), and Distance Intersection Over Union (DIOU) modules, which improves the positive and negative sample imbalance and the problem of small targets being easily missed during model training. Finally, the Region Proposal Network (RPN) is simplified to achieve a faster detection speed and a lower miss rate. The multi-scale training (MST) strategy is also used to train the improved Faster R-CNN to achieve a balance between detection accuracy and efficiency. Compared to the other detection models, the improved Faster R-CNN demonstrates significant advantages in terms of mAP@0.5, F1-score, and Log average miss rate (LAMR). The model proposed in this paper provides valuable insights and inspiration for many fields, such as smart agriculture, medical diagnosis, and face recognition.

Development of a Prototype Solution for Reducing Soup Waste in an Institutional Canteen

Food waste has gained increasing attention and debate, given its economic, environmental, social, and nutritional implications. One-third of food intended for human consumption is wasted. Although it is present at all stages of the food supply chain, it is in the final stages of consumption, such as households and food services, that the problem becomes most evident. This work builds on a previous study by the same authors, which identified computer vision as a suitable technology for identifying and quantifying food waste in institutional canteens. Based on this result, this paper describes the proposal and implementation process of a prototype demonstration. It is based on a Raspberry Pi 4 platform, a ResNet-50 model adapted with the Faster Region-Convolutional Neural Network (Faster R-CNN) model, and an algorithm for feature extracting. A specially built dataset was used to meet the challenge of detecting soup bowls and classifying waste in their consumption. A web application was developed to visualize the data collected, supporting decision making for more efficient food waste management. The prototype was subjected to validation and functional tests, and proved to be a viable, low-cost solution.

Improved faster region convolutional neural network algorithm for UAV target detection in complex environment

With the development of artificial intelligence technology and Unmanned Aerial Vehicle (UAV) technology, the traditional UAV target detection methods are difficult to achieve high detection accuracy in complex and changeable scenes. The paper selects Faster Region Convolutional Neural Network (Faster R-CNN) as the basic algorithm due to its high scalability and excellent classification performance. The convolution mode of the convolution layer in the network structure is adjusted to deformable convolution. Additionally, a hybrid attention mechanism module is added to combine channel attention and spatial attention modules behind the output layer of the Faster R-CNN network. Finally, the Soft Non-Maximum Suppression (Soft-NMS) method is selected to combine the improved Faster R-CNN algorithm of multiple individuals into the final target detection model. The performance of the improved Faster R-CNN algorithm and the original Faster R-CNN algorithm was verified through the VisDrone 2018-DET dataset and the full class average accuracy Mean Average Precision (mAP), accuracy and precision. The accuracy and logarithm of loss values of the improved Faster R-CNN algorithm and the original Faster R-CNN algorithm's Region Proposal Network (RPN) were 0.985, 0.981, 0.018, and 0.052, respectively. The target detection model based on the hybrid attention mechanism of ResNet-50 network fusion Spartial Attention Module (SAM) and Channel Attention Module (CAM) had the best classification performance, with the accuracy, precision and overall average accuracy of 25.8 %, 24.6 % and 22.7 %, respectively. The results are helpful to improve the target detection ability of UAV in complex environment, and contribute to the development of target detection technology in the future.

Proposal-Free Fully Convolutional Network: Object Detection Based on a Box Map.

Region proposal-based detectors, such as Region-Convolutional Neural Networks (R-CNNs), Fast R-CNNs, Faster R-CNNs, and Region-Based Fully Convolutional Networks (R-FCNs), employ a two-stage process involving region proposal generation followed by classification. This approach is effective but computationally intensive and typically slower than proposal-free methods. Therefore, region proposal-free detectors are becoming popular to balance accuracy and speed. This paper proposes a proposal-free, fully convolutional network (PF-FCN) that outperforms other state-of-the-art, proposal-free methods. Unlike traditional region proposal-free methods, PF-FCN can generate a "box map" based on regression training techniques. This box map comprises a set of vectors, each designed to produce bounding boxes corresponding to the positions of objects in the input image. The channel and spatial contextualized sub-network are further designed to learn a "box map". In comparison to renowned proposal-free detectors such as CornerNet, CenterNet, and You Look Only Once (YOLO), PF-FCN utilizes a fully convolutional, single-pass method. By reducing the need for fully connected layers and filtering center points, the method considerably reduces the number of trained parameters and optimizes the scalability across varying input sizes. Evaluations of benchmark datasets suggest the effectiveness of PF-FCN: the proposed model achieved an mAP of 89.6% on PASCAL VOC 2012 and 71.7% on MS COCO, which are higher than those of the baseline Fully Convolutional One-Stage Detector (FCOS) and other classical proposal-free detectors. The results prove the significance of proposal-free detectors in both practical applications and future research.

A Detection Transformer-Based Intelligent Identification Method for Multiple Types of Road Traffic Safety Facilities.

Road traffic safety facilities (TSFs) are of significant importance in the management and maintenance of traffic safety. The complexity and variety of TSFs make it challenging to detect them manually, which renders the work unsustainable. To achieve the objective of automatic TSF detection, a target detection dataset, designated TSF-CQU (TSF data collected by Chongqing University), was constructed based on images collected by a car recorder. This dataset comprises six types of TSFs and 8410 instance samples. A detection transformer with an improved denoising anchor box (DINO) was selected to construct a model that would be suitable for this scenario. For comparison purposes, Faster R-CNN (Region Convolutional Neural Network) and Yolov7 (You Only Look Once version 7) were employed. The DINO model demonstrated the highest performance on the TSF-CQU dataset, with a mean average precision (mAP) of 82.2%. All of the average precision (AP) values exceeded 0.8, except for streetlights (AP = 0.77) and rods (AP = 0.648). The DINO model exhibits minimal instances of erroneous recognition, which substantiates the efficacy of the contrastive denoising training approach. The DINO model rarely makes misjudgments, but a few missed detection.

Breast Cancer Detection in Mammography using Faster Region Convolutional Neural Networks and Group Convolution

Breast cancer is a prevalent and potentially life-threatening medical condition characterized by uncontrolled cell proliferation within breast tissue. This global health concern predominantly among women, although it can affect men as well. The timely and accurate detection of breast cancer is critical, as it significantly influenced treatment outcomes and survival rates. Despite advancements in human-based diagnostic methods, inherent limitations such as subjectivity, human error, and fatigue persist. To address these constraints, computer-vision-based techniques have been explored for breast cancer detection, aiming to enhance early detection, reduce diagnostic errors, and improve patient outcomes. This study advances previous approaches by integrating group convolution and the Special Euclidean motion group as a features extractor into the Faster Region Convolutional Neural Networks framework of Detectron2. This integration offers the advantage of enhancing the Convolutional Neural Network by incorporating a method to preserve invariant and equivariant features, effectively leveraging symmetries in mammography images. The INbreast dataset, comprising 410 images from 115 cases, including 90 instances with bilateral breast involvement in women, was utilized. To ensure data compatibility, DICOM to PNG and XML to JSON format conversions were performed. A data enhancement pipeline, encompassing techniques such as image cropping, truncation normalization, contrast-limited adaptive histogram equalization, and image synthesis, was employed for data preprocessing and cleaning. The proposed technique demonstrated competitive results in breast cancer detection, achieving a recall rate of 97.22. This underscores the efficacy of the integrated approach in improving diagnostic accuracy and holds promise for advancing computer-vision-based breast cancer detection methodologies.

AI- POWERED BASED BLIND AND VISUALLY IMPAIRED SYSTEM FOR SMART GLASS

Visual impairment causes serious problems in daily life and mobility for millions of people worldwide. Against this background, smart assistive technology is gaining momentum to improve the independence and quality of life of the blind and visually impaired. They have difficulty in daily life because they cannot detect the problems around them, and one of their biggest problems is identifying people. Besides automation, there are many unexplored uses of search engines. This project includes an app that uses search to help blind people see things in front of them safely, and a facial recognition system with visual feedback that can help visually impaired people see familiar and unfamiliar people. The speaker will provide them with sound support. In this work, we use deep learning-based Regional Convolutional Neural Network (Faster R-CNN) to detect and identify people and objects in the environment. Faster local convolutional neural network technology processes and distributes images captured by cameras. The operator takes the image as input. Therefore, this model helps visually impaired people in a simpler way than a white cane. Keywords—Visuallmpairment,Objectdetection,Faster Region Convolutional Neural Network.

Defect Detection of Cell Phone Screen Using a Faster Regional Convolutional Neural Network with Multi-head Attention Mechanism

Since the glass outer screen of a cell phone is the main sensory part of the human eye when using a cell phone, the advantages and disadvantages of the cell phone screen directly affect people's sense of use. Therefore, the defect detection requirements for cell phone screens are high and need to meet the needs of high-volume factory inspection. Most of the traditional defect detection methods use visual methods, the detection results are overly dependent on the subjectivity and experience of workers, the efficiency of this method is low, and the accuracy is poor. Currently, machine learning-based detection methods are applied in numerous industries. In this paper, a faster Regional Convolutional Neural Network (R-CNN) with multi-head attention mechanism for defect detection of cell phone screen is proposed. To enhance the network's capability in extracting feature information, a four-head attention mechanism is added to the last convolutional layer of the ResNet50 network. An improved Region of Interest (ROI) Align is proposed to replace the original ROI Pooling to reduce the localization error of cell phone screen defects. Replace the original Rectified Linear Unit (ReLU) activation function with the Copy Exponential Linear Unit (CELU) activation function to expedite the convergence capability of the network. Finally, by comparing with other classical model training, the evaluation results indicate that the proposed method achieved an average accuracy of 95.71%, which is a 5.34% improvement compared to the original faster R-CNN network.

A deep learning model based on magnifying endoscopy with narrow-band imaging to evaluate intestinal metaplasia grading and OLGIM staging: A multicenter study

Background and PurposePatients with stage III or IV of operative link for gastric intestinal metaplasia assessment (OLGIM) are at a higher risk of gastric cancer (GC). We aimed to construct a deep learning (DL) model based on magnifying endoscopy with narrow-band imaging (ME-NBI) to evaluate OLGIM staging. MethodsThis study included 4473 ME-NBI images obtained from 803 patients at three endoscopy centres. The endoscopic expert marked intestinal metaplasia (IM) regions on endoscopic images of the target biopsy sites. Faster Region-Convolutional Neural Network model was used to grade IM lesions and predict OLGIM staging. ResultsThe diagnostic performance of the model for IM grading in internal and external validation sets, as measured by the area under the curve (AUC), was 0.872 and 0.803, respectively. The accuracy of this model in predicting the high-risk stage of OLGIM was 84.0%, which was not statistically different from that of three junior (71.3%, p = 0.148) and three senior endoscopists (75.3%, p = 0.317) specially trained in endoscopic images corresponding to pathological IM grade, but higher than that of three untrained junior endoscopists (64.0%, p = 0.023). ConclusionThis DL model can assist endoscopists in predicting OLGIM staging using ME-NBI without biopsy, thereby facilitating screening high-risk patients for GC.

Faster Region Convolutional Neural Network (FRCNN) Based Facial Emotion Recognition

Faster Region Convolutional Neural Network (FRCNN) Based Facial Emotion Recognition

Classification and Localization of Arabic Handwritten Text in KHATT Dataset Based on Faster R-CNN

Arabic handwriting recognition is an important research area in computer vision. Due to the complexity of the Arabic script, this is an arduous task. Several approaches have been proposed to address this challenge, including deep learning algorithms. Despite their efficiency, these approaches present some limitations such as the use of lexicon-driven models, the need of a lot of data for training and the huge computational cost. We propose, in this paper, two novel models based on the robust Faster Region-Convolution Neural Network (Faster R-CNN) to recognize Arabic handwritten sentences. The Faster R-CNN is commonly used to detect objects in images and depends on region proposal algorithms. These models use the pre-trained VGG 16 and ResNet50. Moreover, they use a soft non-maximum suppression strategy (Soft-NMS) in the post-processing stage instead of the traditional NMS to increase the detection rate of overlapping items. The models will be trained independently to identify words in the text lines and tested using sentences from the (KFUPM Handwritten Arabic TexT) KHATT dataset. It will be shown that the faster R-CNN models allow greater accuracy and effectiveness in handwriting recognition. They achieve accuracy rates of 99% and 98% for the two networks, VGG16 and ResNet50, respectively

Melanoma Diagnosis Using Enhanced Faster Region Convolutional Neural Networks Optimized by Artificial Gorilla Troops Algorithm

Melanoma, a rapidly spreading and perilous type of skin cancer, is the focus of this study, presenting a reliable technique for its detection. It is one of the most prevalent types of cancer that might be challenging for medical professionals to diagnose. Artificial intelligence can improve diagnostic accuracy when utilized in conjunction with the expertise of medical specialists. An innovative computer-aided method for the diagnosis of skin cancer has been introduced in the current study. The construction of the proposed method uses the African Gorilla Troops Optimizer (AGTO) Algorithm, a recently introduced meta-heuristic optimization algorithm, and deep learning models such as Faster Region Convolutional Neural Networks. To reduce the complexity of the analytic process, valuable features are chosen using the AGTO method, and further classification is implemented using Faster R-CNN. The proposed model is applied to the ISIC-2020 skin cancer dataset. When the final performance results from the proposed model are compared to those from four existing works, the findings show that the proposed system outperforms the existing models with an accuracy of 98.55%.

Image Object Detection Method Based on Improved Faster R-CNN

Aiming at the problems about the incompleteness and low accuracy in classifying image feature extraction in existing image object detection methods. First, an improved image object detection method based on faster Region-Convolutional Neural Network (RCNN) is proposed, and Region of Interest (ROI) align is used instead of RoI pooling to reduce the error in the pooling process. Second, it adds the layer of convolution before adding network full-connection layer, thus reducing its parameters, enhance the performance of the classifier and avoid over-fitting. Finally, the combination of softmax and central loss functions is for training network-based model to increase the differences between categories and reduce the changes within categories, and thus the model with network could have more targeted and diversified features. Research results found that in total, method average accuracy is 91.04%, indicating higher than the accuracy of other methods and has good image target detection ability.

Microscopic hyperspectral imaging and an improved detection model based detection of Mycogone perniciosa chlamydospore in soil

The Mycogone perniciosa disease of Agaricus bisporus is highly contagious, with insignificant early symptoms and a long infestation period. Currently, there is a lack of a convenient and rapid means to detect the disease for early control. By using the transmission route of the disease, this paper focuses on M. perniciosa chlamydospore detection in contaminated soil. Microscopic hyperspectral images of M. perniciosa chlamydospore in contaminated soil were obtained to establish a detection model. Given a small target and a complex background, we proposed an improved spore detection model based on a faster regional convolutional neural network (Faster R-CNN). Furthermore, we combined the residual network Resnet50 and feature pyramid network (FPN) to extract thick spore target features at multiple scales. Meanwhile, we optimized the region proposal network (RPN) region proposal generation by adding two small scales to improve the performance of the detection model. The first three principal components (with 95% or more information) and RGB images were selected as model inputs, respectively, and the final average precision (AP) was 94.68% and 92.35%, respectively. This PC-based model also was compared to the VGG16 and Resnet50-based feature extraction networks of Faster R-CNN and Darknet53-based feature extraction network of the YOLOv3 model, and the AP was found to improve by 5.41%, 4.78%, and 6.34%, respectively. The results showed that micro-hyperspectral imaging combined with deep learning methods could accurately detect the chlamydospore in the soil, providing new methods and ideas for the early prevention and detection of M. perniciosa disease of A. bisporus.

Online verification and management scheme of gateway meter flow in the power system by machine learning

Currently, the calibration of electric energy meters often involves manual meter reading, dismantling inspection, or regular sampling inspection conducted by professionals. To improve work efficiency and verification accuracy, this research integrates machine learning into the scheme of online verification and management of gateway meter flow in the power system. The approach begins by applying the Faster Region Convolutional Neural Network (Faster-RCNN) model and the Single Shot MultiBox Detector (SSD) model to the recognition system for dial readings. Then, the collected measurement data is pre-processed, excluding data collected under light load conditions. Next, an estimation error model and a solution equation for the electricity meter are established based on the pre-processed data. The operation error of the electricity meter is estimated, and the estimation accuracy is verified using the limited memory recursive least squares algorithm (LMRLSA). Furthermore, business assistant decision-making is carried out by combining the remote verification results with the estimation outcomes. The proposed dial reading recognition system is tested using 528 images of meter readings, achieving an accuracy of 98.49%. In addition, the influence of various parameters on the error results of the electricity meter is also explored. The results demonstrate that a memory length ranging from 600 to 1,200 and a line loss error of less than 5% yield the most suitable accuracy for estimating the electricity meter error. Meanwhile, it is advisable to remove measurement data collected under light load to avoid unnecessary checks. The experiments manifest that the proposed algorithm can properly eliminate the influence of old measurement data on the error parameter estimation, thereby enhancing the accuracy of the estimation. The adjustment of the memory length ensures real-time performance in estimating meter errors and enables online monitoring. This research has certain reference significance for achieving the online verification and management of gateway meter flow in the power system.

E-YOLO: Recognition of estrus cow based on improved YOLOv8n model

Timely and accurately identifying estrus cows is essential in modern dairy farming. To address the challenges of delayed and inefficient manual monitoring of cow estrus, an improved model based on You Only Look Once v8 Nano (YOLOv8n), named Estrus-YOLO (E-YOLO), was proposed to identify estrus cows efficiently. In this research, the dataset was labelled not only for cow mounting behavior but also innovatively labelled individual estrus cows, enabling precise identification of estrus cows. Due to the small size of cow in the field of view, the Complete Intersection over Union (CIoU) loss was replaced with the Normalized Wasserstein Distance (NWD) loss to reduce sensitivity to position deviations of target cows. Context Information Augmentation Module (CIAM) was proposed to enhance the contextual information for estrus cows by utilizing cow mounting behavior as reference features. Furthermore, the Triplet Attention Module (TAM) was incorporated into the Backbone to enhance the network’s focus on individual estrus cows through cross-dimensional interactions. To validate the effectiveness of the algorithm, experiments were conducted on a dataset consisting of 1716 instances of cow mounting behavior. The experimental results demonstrated that the proposed model achieved an Average Precision of estrus (APestrus) of 93.90%, Average Precision of mounting (APmounting) of 95.70%, F1-score of 93.74%, detection speed of 8.1 ms/frame, with the parameters of 3.04 M, and the Floating-point Operations (FLOPs) of 9.9 G. Compared to the YOLOv8 model, the proposed model exhibited an improvement of 5.40% in APestrus and 3.30% in APmounting. When compared to Single Shot MultiBox Detector (SSD), Faster Region Convolutional Neural Network (Faster R-CNN), YOLOv5n, YOLOv5s, YOLOv7-tiny, YOLOv8n, and YOLOv8s, the proposed model had fewer parameters, FLOPs, and fast detection speed. Except for APmounting, which was slightly lower than SSD, the rest accuracy indexes were the highest, showing good comprehensive performance and meeting the requirements of accurate and rapid identification of estrus cows. The proposed model was helpful for accurate and real-time monitoring of estrus cows in complex breeding environments and all-weather conditions.