Spike Detection Research Articles

Neuronal Multi Unit Activity Processing with Metal Oxide Memristive Devices

AbstractIntra‐cortical brain‐machine interfaces (BMIs), able to decode neural activity in real‐time, represent a revolutionary opportunity for treating medical conditions. However, traditional systems focusing on single‐neuron spike detection require high processing rates and power, hindering the up‐scaling for neurons‐population monitoring in clinical application. An intriguing proposition is the memristive integrating sensor (MIS) approach, which uses resistive RAM (RRAM) for threshold‐based neural activity detection. MIS leverages analogue multi‐state switching properties of metal‐oxide RRAM to compress neural inputs by encoding above‐threshold events in resistance displacement, facilitating efficient data down‐sampling in the post‐processing, enabling low‐power, high‐channel systems. Initially tested on spikes and local field potentials, here MIS is adapted to process multi‐unit activity envelope (eMUA)—the envelope of entire spiking activity—which has recently been proposed as crucial input for real‐time neuro‐prosthetic control. Prior necessary modifications to the MIS for effective operation, this adaptation achieved over 95% sensitivity across two types of metal‐oxide devices: Pt/TiOx/Pt and TiN/HfOx/TiN, proving its platform‐agnostic capabilities. Furthermore, towards the integration of MIS with silicon chips, it is shown that it can reduce total system power consumption to below 1 µW, as RRAM encoding stage relaxes the signal preservation and noise requirements that challenge traditional complementary metal‐oxide‐semiconductor (CMOS) front‐ends. This eMUA‐MIS adaptation offers a viable pathway for developing more scalable and efficient BMIs for clinical use.

Deep learning-based spike sorting: a survey.

Objective.Deep learning is increasingly permeating neuroscience, leading to a rise in signal-processing applications for extracellular recordings. These signals capture the activity of small neuronal populations, necessitating 'spike sorting' to assign action potentials (spikes) to their underlying neurons. With the rise in publications delving into new methodologies and techniques for deep learning-based spike sorting, it is crucial to synthesise these findings critically. This survey provides an in-depth evaluation of the approaches, methodologies and outcomes presented in recent articles, shedding light on the current state-of-the-art.Approach.Twenty-four articles published until December 2023 on deep learning-based spike sorting have been examined. The proposed methods are divided into three sub-problems of spike sorting: spike detection, feature extraction and classification. Moreover, integrated systems, i.e., models that detect spikes and extract features or do classification within a single network, are included.Main Results.Although most algorithms have been developed for single-channel recordings, models utilising multi-channel data have already shown promising results, with efficient hardware implementations running quantised models on ASICs and FPGAs. Convolutional neural networks have been used extensively for spike detection and classification as the data can be processed spatiotemporally while maintaining low-parameter models and increasing generalisation and efficiency. Autoencoders have been mainly utilised for dimensionality reduction, enabling subsequent clustering with standard methods. Also, integrated systems have shown great potential in solving the spike sorting problem from end to end.Significance.This survey explores recent articles on deep learning-based spike sorting and highlights the capabilities of deep neural networks in overcoming associated challenges, but also highlights potential biases of certain models. Serving as a resource for both newcomers and seasoned researchers in the field, this work provides insights into the latest advancements and may inspire future model development.

Multimodal and quantitative analysis of the epileptogenic zone network in the pre-surgical evaluation of drug-resistant focal epilepsy

Surgical resection for epilepsy often fails due to incomplete Epileptogenic Zone Network (EZN) localization from scalp electroencephalography (EEG), stereo-EEG (SEEG), and Magnetic Resonance Imaging (MRI). Subjective interpretation based on interictal, or ictal recordings limits conventional EZN localization. This study employs multimodal analysis using high-density-EEG (HDEEG), Magnetoencephalography (MEG), functional-MRI (fMRI), and SEEG to overcome these limitations in a patient with drug-resistant MRI-negative focal epilepsy. A 17-year-old with drug-resistant epilepsy underwent evaluation. HDEEG, MEG, fMRI, and SEEG were used, with a novel HDEEG-cap facilitating simultaneous EEG-MEG and EEG-fMRI recordings. Electrical and magnetic source imaging were performed, and fMRI data were analysed for homogenous regions. SEEG analysis involved spike detection, spike timing analysis, ictal fast activity quantification, and Granger-based connectivity analysis. Non-invasive sessions revealed consistent interictal source imaging results identifying the EZN in the right anterior cingulate cortex. EEG-fMRI highlighted broader activation in the right cingulate cortex. SEEG analysis localized spikes and fast activity in the right anterior and posterior cingulate gyri. Multi-modal analysis suggested the EZN in the right frontal lobe, primarily involving the anterior and mid-cingulate cortices. Multi-modal non-invasive analyses can optimise SEEG implantation and surgical decision-making. Invasive analyses corroborated non-invasive findings, emphasising the importance of individual-case quantitative analysis across modalities in complex epilepsy cases.

Deep learning based automatic detection and dipole estimation of epileptic discharges in MEG: a multi-center study.

Magnetoencephalography (MEG) provides crucial information in diagnosing focal epilepsy. However, dipole estimation, a commonly used analysis method for MEG, can be time-consuming since it necessitates neurophysiologists to manually identify epileptic spikes. To reduce this burden, we developed the automatic detection of spikes using deep learning in single center. In this study, we performed a multi-center study using six MEG centers to improve the performance of the automated detection of neuromagnetically recorded epileptic spikes, which we previously developed using deep learning. Data from four centers were used for training and evaluation (internal data), and the remaining two centers were used for evaluation only (external data). We used a five-fold subject-wise cross-validation technique to train and evaluate the models. A comparison showed that the multi-center model outperformed the single-center model in terms of performance. The multi-center model achieved an average ROC-AUC of 0.9929 and 0.9426 for the internal and external data, respectively. The median distance between the neurophysiologist-analyzed and automatically analyzed dipoles was 4.36 and 7.23mm for the multi-center model for internal and external data, respectively, indicating accurate detection of epileptic spikes. By training data from multiple centers, automated analysis can improve spike detection and reduce the analysis workload for neurophysiologists. This study suggests that the multi-center model has the potential to detect spikes within 1cm of a neurophysiologist's analysis. This multi-center model can significantly reduce the number of hours required by neurophysiologists to detect spikes.

A generalized model for accurate wheat spike detection and counting in complex scenarios

Wheat is a crucial crop worldwide, and accurate detection and counting of wheat spikes are vital for yield estimation and breeding. However, these tasks are daunting in complex field environments. To tackle this, we introduce RIA-SpikeNet, a model designed to detect and count wheat spikes in such conditions. First, we introduce an Implicit Decoupling Detection Head to incorporate more implicit knowledge, enabling the model to better distinguish visually similar wheat spikes. Second, Asymmetric Loss is employed as the confidence loss function, enhancing the learning weights of positive and hard samples, thus improving performance in complex scenes. Lastly, the backbone network is modified through reparameterization and the use of larger convolutional kernels, expanding the effective receptive field and improving shape information extraction. These enhancements significantly improve the model’s ability to detect and count wheat spikes accurately. RIA-SpikeNet outperforms the state-of-the-art YOLOv8 detection model, achieving a competitive 81.54% mAP and 90.29% R2. The model demonstrates superior performance in challenging scenarios, providing an effective tool for wheat spike yield estimation in field environments and valuable support for wheat production and breeding efforts.

Feature diffusion reconstruction mechanism network for crop spike head detection.

Monitoring crop spike growth using low-altitude remote sensing images is essential for precision agriculture, as it enables accurate crop health assessment and yield estimation. Despite the advancements in deep learning-based visual recognition, existing crop spike detection methods struggle to balance computational efficiency with accuracy in complex multi-scale environments, particularly on resource-constrained low-altitude remote sensing platforms. To address this gap, we propose FDRMNet, a novel feature diffusion reconstruction mechanism network designed to accurately detect crop spikes in challenging scenarios. The core innovation of FDRMNet lies in its multi-scale feature focus reconstruction and lightweight parameter-sharing detection head, which can effectively improve the computational efficiency of the model while enhancing the model's ability to perceive spike shape and texture.FDRMNet introduces a Multi-Scale Feature Focus Reconstruction module that integrates feature information across different scales and employs various convolutional kernels to capture global context effectively. Additionally, an Attention-Enhanced Feature Fusion Module is developed to improve the interaction between different feature map positions, leveraging adaptive average pooling and convolution operations to enhance the model's focus on critical features. To ensure suitability for low-altitude platforms with limited computational resources, we incorporate a Lightweight Parameter Sharing Detection Head, which reduces the model's parameter count by sharing weights across convolutional layers. According to the evaluation experiments on the global wheat head detection dataset and diverse rice panicle detection dataset, FDRMNet outperforms other state-of-the-art methods with mAP@.5 of 94.23%, 75.13% and R 2 value of 0.969, 0.963 between predicted values and ground truth values. In addition, the model's frames per second and parameters in the two datasets are 227.27,288 and 6.8M, respectively, which maintains the top three position among all the compared algorithms. Extensive qualitative and quantitative experiments demonstrate that FDRMNet significantly outperforms existing methods in spike detection and counting tasks, achieving higher detection accuracy with lower computational complexity.The results underscore the model's superior practicality and generalization capability in real-world applications. This research contributes a highly efficient and computationally effective solution for crop spike detection, offering substantial benefits to precision agriculture practices.

SynchroLINNce: toolbox for Neural Synchronization and Desynchronization Assessment in Epilepsy Animal Models

Epilepsy is a worldwide public health issue, given its biological, social, and economic impacts. Considering several open questions about synchronization and desynchronization mechanisms underlying epileptic phenomena, the development of algorithms and computational toolboxes for such analysis is highly relevant to their research. Moreover, given the recent developments of neurotechnology for epilepsy, it is essential to understand that proposals like computational tools may provide consistent data for closed-loop control systems, necessary in neuromodulation treatment alternatives, and for real-time monitoring systems to predict the occurrence of epileptic seizures. In the present work, SynchroLINNce, a freely distributable MATLAB toolbox designed to be used by epilepsy neuroscientists, including software-untrained), is proposed. Among its features, several functionalities such as recording visualization, digital filtering, and correlation analysis, as well as more specific methodologies, such as mechanisms for the automatic detection of epileptiform spikes, morphology analysis of these spikes, and their coincidence between channels are presented.

Lightweight and Efficient Wheat Spike Detection for Small Targets

Lightweight and Efficient Wheat Spike Detection for Small Targets

Clinicopathological characteristics of neural epidermal growth factor-like 1 protein-associated membranous glomerulonephritis.

Neural epidermal growth factor-like 1 protein (NELL1) is the second most common target antigen in membranous glomerulonephritis (MGN). However, data regarding the clinicopathological characteristics of NELL1-associated MGN are limited owing to its low prevalence. This study examined the prevalence and clinicopathological characteristics of NELL1-associated MGN in a Japanese cohort. Additionally, we compared the clinicopathological features of NELL1-positive MGN, phospholipase A2 receptor 1 (PLA2R1)-positive MGN, and MGN negative for all three antigens (NELL1, PLA2R1, and thrombospondin type-1 domain-containing 7A). Among 257 consecutive patients pathologically diagnosed with MGN at two centers in Japan, 24 (9.3%) were immunohistochemically positive for NELL1. Clinically, patients with NELL1-positive MGN were significantly older (p < 0.001) and had a higher frequency of bucillamine use (vs PLA2R1-positive MGN, p < 0.01). Histologically, NELL1-positive MGN exhibited significantly lower detection of spikes and crater formation (p < 0.001), higher prevalence of segmental spike distribution (vs PLA2R1-positive MGN: p < 0.001), and higher prevalence of stage I cases on electron microscopy (p < 0.01). There were no significant differences in the prognoses among the three groups. The characteristic histological feature of segmental distribution in NELL1-positive MGN may be related to bucillamine use and the early phase of the disease. Further investigations with larger numbers of patients may offer further insight into the prognosis of patients with NELL1-positive MGN.

A novel spike detection model for dynamic stress monitoring of bogie frame

The fatigue evaluation of the bogie frame is an important part of the structural health monitoring of the vehicle. During the dynamic stress monitoring, some signal spikes, which are much larger than the normal fluctuation range due to the interference of the complex electromagnetic environment, affect the accuracy of the structural damage assessment and need to be accurately detected and replaced. Aiming at the drawbacks of traditional detection methods that are overly dependent on engineering experience and not universal, a novel spike detection model is proposed in this paper. By the process of data transformation, spike region features are effectively separated. Based on the isolation forest algorithm, the normalized anomaly score of each point is calculated, and the threshold is determined adaptively. The spike detection rate and damage sensitivity are proposed as the evaluation indices of the detection effect of the method. The results show that the spike detection rate is improved by 7.86% on average, and the damage sensitivity is improved by 15.59% on average. The spike detection model in this paper is significantly improved compared to the existing methods.

A high‐performance asynchronous readout circuit with cascade function for a neural recording micro‐electrode array

AbstractThe data‐driven machine learning technology used for neural decoding emphasizes the requirements of in vivo and in vitro neural signal acquisition with high spatial and temporal resolution. However, micro‐electrode arrays (MEAs) that simultaneously achieve high spatial and temporal resolution neural signal acquisition are not yet available. Meanwhile, the high data bandwidth of large‐scale MEA brings challenges in power consumption, data transmission, storage, and neural signal processing. This research aims to improve the spatial and temporal resolution of MEA, reduce the cost and time of large‐scale MEA customization through cascading, and reduce the bandwidth of large‐scale MEA through spike compression. Firstly, based on in‐pixel spike detection, a row‐based neural spike readout mechanism and related array circuit to improve the neural spike readout performance and temporal resolution of neural signal acquisition is proposed. To further enhance the spatial resolution and reduce the risk of large‐scale MEA fabrication, the cascading capability of the readout circuit is explored. Lastly, spatial correlation‐based neural spike encoding is proposed to reduce the data bandwidth, achieving a 5.2× compression rate. This is a study on implementing large‐scale MEA through cascaded readout circuits and novel study to utilize the spatial correlation between detected neural spikes for further compression.

SNSDeepNet: spike and non-spike detection in epilepsy

Epilepsy, a severe neurological condition is marked by sharp waveforms known as spikes in electroencephalogram (EEG) signals. Prompt detection of these spikes is crucial for reducing accidental injuries and safeguarding the lives of epilepsy patients. This article proposes an innovative deep-learning approach for epileptic spike detection using Spike and Non-spike Deep Convolutional Neural Networks (SNSDeepNet). Our method utilizes CNNs alongside an adaptive Layer-wise Adaptive Moments (LAMB) optimizer to effectively extract relevant features from time-domain (TD) and frequency-domain (FD) representations of spike and non-spike signals. The adaptive LAMB optimizer enhances the training process and accelerates convergence compared to traditional optimizers. The proposed model is evaluated using EEG recordings from three datasets: the Children’s Hospital Boston (CHB-MIT) dataset, the Siena Scalp EEG dataset (Physionet Siena Scalp EEG Database), and the Bonn EEG dataset from the University of Bonn. After pre-processing and applying a peak detection algorithm, we extract TD and FD features from the signals. Our model demonstrates impressive performance. The CHB-MIT dataset achieved an average accuracy of 99.69%, sensitivity of 99.68%, F1-score of 99.11%, and a false positive rate (FPR) of 0.026 98. For the Siena dataset, the model achieved an accuracy of 99.62%, specificity of 99.04%, sensitivity of 99.93%, F1-score of 99.48%, and an FPR of 0.009 208. The Bonn dataset achieved an average accuracy of 94.10%, specificity of 92.39%, sensitivity of 97.35%, and an FPR of 0.0764. These findings underscore the effectiveness of the proposed architecture in accurately identifying epileptic spikes, highlighting its potential to enhance epilepsy diagnosis and treatment.

Real-Time Detection and Counting of Wheat Spikes Based on Improved YOLOv10

Wheat is one of the most crucial food crops globally, with its yield directly impacting global food security. The accurate detection and counting of wheat spikes is essential for monitoring wheat growth, predicting yield, and managing fields. However, the current methods face challenges, such as spike size variation, shading, weed interference, and dense distribution. Conventional machine learning approaches have partially addressed these challenges, yet they are hampered by limited detection accuracy, complexities in feature extraction, and poor robustness under complex field conditions. In this paper, we propose an improved YOLOv10 algorithm that significantly enhances the model’s feature extraction and detection capabilities. This is achieved by introducing a bidirectional feature pyramid network (BiFPN), a separated and enhancement attention module (SEAM), and a global context network (GCNet). BiFPN leverages both top-down and bottom-up bidirectional paths to achieve multi-scale feature fusion, improving performance in detecting targets of various scales. SEAM enhances feature representation quality and model performance in complex environments by separately augmenting the attention mechanism for channel and spatial features. GCNet captures long-range dependencies in the image through the global context block, enabling the model to process complex information more accurately. The experimental results demonstrate that our method achieved a precision of 93.69%, a recall of 91.70%, and a mean average precision (mAP) of 95.10% in wheat spike detection, outperforming the benchmark YOLOv10 model by 2.02% in precision, 2.92% in recall, and 1.56% in mAP. Additionally, the coefficient of determination (R2) between the detected and manually counted wheat spikes was 0.96, with a mean absolute error (MAE) of 3.57 and a root-mean-square error (RMSE) of 4.09, indicating strong correlation and high accuracy. The improved YOLOv10 algorithm effectively solves the difficult problem of wheat spike detection under complex field conditions, providing strong support for agricultural production and research.

YOLO-LF: a lightweight multi-scale feature fusion algorithm for wheat spike detection

YOLO-LF: a lightweight multi-scale feature fusion algorithm for wheat spike detection

Prediction of Feed Quantity for Wheat Combine Harvester Based on Improved YOLOv5s and Weight of Single Wheat Plant without Stubble

In complex field environments, wheat grows densely with overlapping organs and different plant weights. It is difficult to accurately predict feed quantity for wheat combine harvester using the existing YOLOv5s and uniform weight of a single wheat plant in a whole field. This paper proposes a feed quantity prediction method based on the improved YOLOv5s and weight of a single wheat plant without stubble. The improved YOLOv5s optimizes Backbone with compact bases to enhance wheat spike detection and reduce computational redundancy. The Neck incorporates a hierarchical residual module to enhance YOLOv5s’ representation of multi-scale features. The Head enhances the detection accuracy of small, dense wheat spikes in a large field of view. In addition, the height of a single wheat plant without stubble is estimated by the depth distribution of the wheat spike region and stubble height. The relationship model between the height and weight of a single wheat plant without stubble is fitted by experiments. Then, feed quantity can be predicted using the weight of a single wheat plant without stubble estimated by the relationship model and the number of wheat plants detected by the improved YOLOv5s. The proposed method was verified through experiments with the 4LZ-6A combine harvester. Compared with the existing YOLOv5s, YOLOv7, SSD, Faster R-CNN, and other enhancements in this paper, the mAP50 of wheat spikes detection by the improved YOLOv5s increased by over 6.8%. It achieved an average relative error of 4.19% with a prediction time of 1.34 s. The proposed method can accurately and rapidly predict feed quantity for wheat combine harvesters and further realize closed-loop control of intelligent harvesting operations.

A data compression algorithm with the improved SRLE for high-throughput neural signal acquisition device.

Multi-channel acquisition systems of brain neural signals can provide a powerful tool with a wide range of information for the clinical application of brain computer interfaces. High-throughput implantable systems are limited by size and power consumption, posing challenges to system design. To acquire more comprehensive neural signals and wirelessly transmit high-throughput brain neural signals, a FPGA-based acquisition system for multi-channel brain nerve signals has been developed. And the Bluetooth transmission with low-power technology are utilized. To wirelessly transmit large amount of data with limited Bluetooth bandwidth and improve the accuracy of neural signal decoding, an improved sharing run length encoding (SRLE) is proposed to compress the spike data of brain neural signal to improve the transmission efficiency of the system. The functional prototype has been developed, which consists of multi-channel data acquisition chips, FPGA main control module with the improved SRLE, a wireless data transmitter, a wireless data receiver and an upper computer. And the developed functional prototype was tested for spike detection of brain neural signal by animal experiments. From the animal experiments, it shows that the system can successfully collect and transmit brain nerve signals. And the improved SRLE algorithm has an excellent compression effect with the average compression rate of 5.94%, compared to the double run-length encoding, the FDR encoding, and the traditional run-length encoding. The developed system, incorporating the improved SRLE algorithm, is capable of wirelessly capturing spike signals with 1024 channels, thereby realizing the implantable systems of High-throughput brain neural signals.

A comprehensive review of spike sorting algorithms in neuroscience

Spike sorting plays a pivotal role in neuroscience, serving as a crucial step of separating electrical signals recorded from multiple neurons to further analyze neuronal interactions. This process involves separating electrical signals that originate from multiple neurons, recorded through devices like electrode arrays. This is a very important link in the field of brain-computer interfaces. The objective of spike sorting algorithm (SSA) is to distinguish the behavior of one or more neurons from background noise using the waveforms captured by brain-embedded electrodes. This article starts from the steps of the conventional SSA and divides the SSA into three steps: spike detection, spike feature extraction, and spike clustering. It outlines prevalent algorithms for each phase before delving into two emerging technologies: template matching and deep learning-based methods. The discussion on deep learning is further subdivided into three approaches: end-to-end solution, deep learning for spike sorting steps, and spiking neural networks-based solutions. Finally, it elaborates future challenges and development trends of SSAs.

Sensitivity-improved SERS detection of SARS-CoV-2 spike protein by Au NPs/COFs integrated with catalytic-hairpin-assembly amplification technology

BackgroundThe COVID-19 pandemic, caused by the novel coronavirus, has had a profound impact on global health and economies worldwide. This unprecedented crisis has affected individuals, communities, and nations in diverse manners. Developing simple and accurate diagnostic methods is an imperative task for frequent testing to mitigate the spread of the virus. Among these methods, SARS-CoV-2 antigen tests in clinical specimens have emerged as a promising diagnostic method for COVID-19 due to their sensitive and accurate detection of spike (S) protein, which plays a crucial role in viral infection initiation. ResultsIn this work, a dual-signal amplification surface enhanced Raman scattering (SERS)-based S protein biosensor was constructed based on Au NPs/COFs and enzyme-free catalytic hairpin assembly (CHA) amplification method. The approach relies on a released free DNA sequence (T), which is generated from the competition reaction between Aptamer/T and Aptamer/S protein, to trigger a CHA reaction. Due to the high binding affinity and selectivity between the S protein and its aptamer, CHA process was triggered with the maximum SERS tags (H2-conjugated Au@4-mercaptobenzonitrile@Ag) anchored onto Au NPs/COFs substrate surface. This SERS platform could detect the S protein at concentrations with high sensitivity (limit of detection = 3.0 × 10−16 g/mL), wide detection range (1 × 10−16 to 1 × 10−11 g/mL), acceptable reproducibility (relative standard deviation = 7.01 %) and excellent specificity. The biosensor was also employed to detect S protein in artificial human salivas. SignificanceThus, this study not only developed a novel Au NPs/COFs substrate exhibiting strong SERS enhancement ability and high reproducibility, but also proposed a promising dual-signal amplification SERS-based diagnostic method for COVID-19, holding immense potential for the detection of a wide range of antigens and infectious diseases in future applications.

Improved YOLO-FastestV2 wheat spike detection model based on a multi-stage attention mechanism with a LightFPN detection head.

The number of wheat spikes has an important influence on wheat yield, and the rapid and accurate detection of wheat spike numbers is of great significance for wheat yield estimation and food security. Computer vision and machine learning have been widely studied as potential alternatives to human detection. However, models with high accuracy are computationally intensive and time consuming, and lightweight models tend to have lower precision. To address these concerns, YOLO-FastestV2 was selected as the base model for the comprehensive study and analysis of wheat sheaf detection. In this study, we constructed a wheat target detection dataset comprising 11,451 images and 496,974 bounding boxes. The dataset for this study was constructed based on the Global Wheat Detection Dataset and the Wheat Sheaf Detection Dataset, which was published by PP Flying Paddle. We selected three attention mechanisms, Large Separable Kernel Attention (LSKA), Efficient Channel Attention (ECA), and Efficient Multi-Scale Attention (EMA), to enhance the feature extraction capability of the backbone network and improve the accuracy of the underlying model. First, the attention mechanism was added after the base and output phases of the backbone network. Second, the attention mechanism that further improved the model accuracy after the base and output phases was selected to construct the model with a two-phase added attention mechanism. On the other hand, we constructed SimLightFPN to improve the model accuracy by introducing SimConv to improve the LightFPN module. The results of the study showed that the YOLO-FastestV2-SimLightFPN-ECA-EMA hybrid model, which incorporates the ECA attention mechanism in the base stage and introduces the EMA attention mechanism and the combination of SimLightFPN modules in the output stage, has the best overall performance. The accuracy of the model was P=83.91%, R=78.35%, AP= 81.52%, and F1 = 81.03%, and it ranked first in the GPI (0.84) in the overall evaluation. The research examines the deployment of wheat ear detection and counting models on devices with constrained resources, delivering novel solutions for the evolution of agricultural automation and precision agriculture.

WH-DETR: An Efficient Network Architecture for Wheat Spike Detection in Complex Backgrounds

Wheat spike detection is crucial for estimating wheat yields and has a significant impact on the modernization of wheat cultivation and the advancement of precision agriculture. This study explores the application of the DETR (Detection Transformer) architecture in wheat spike detection, introducing a new perspective to this task. We propose a high-precision end-to-end network named WH-DETR, which is based on an enhanced RT-DETR architecture. Initially, we employ data augmentation techniques such as image rotation, scaling, and random occlusion on the GWHD2021 dataset to improve the model’s generalization across various scenarios. A lightweight feature pyramid, GS-BiFPN, is implemented in the network’s neck section to effectively extract the multi-scale features of wheat spikes in complex environments, such as those with occlusions, overlaps, and extreme lighting conditions. Additionally, the introduction of GSConv enhances the network precision while reducing the computational costs, thereby controlling the detection speed. Furthermore, the EIoU metric is integrated into the loss function, refined to better focus on partially occluded or overlapping spikes. The testing results on the dataset demonstrate that this method achieves an Average Precision (AP) of 95.7%, surpassing current state-of-the-art object detection methods in both precision and speed. These findings confirm that our approach more closely meets the practical requirements for wheat spike detection compared to existing methods.