Improved YOLO v5 Wheat Ear Detection Algorithm Based on Attention Mechanism

Parkinson's image detection and classification based on deep learning

OBJECTIVE

There are two major issues in the MRI image diagnosis task for Parkinson's disease. Firstly, there are slight differences in MRI images between healthy individuals and Parkinson's patients, and the medical field has not yet established precise lesion localization standards, which poses a huge challenge for the effective prediction of Parkinson's disease through MRI images. Secondly, the early diagnosis of Parkinson's disease traditionally relies on the subjective judgment of doctors, which leads to insufficient accuracy and consistency. This article proposes an improved YOLOv5 detection algorithm based on deep learning for predicting and classifying Parkinson's images.

METHODS

This article improves the YOLOv5s network as the basic framework. Firstly, the CA attention mechanism was introduced to enable the model to dynamically adjust attention based on local features of the image, significantly enhancing the sensitivity of the model to PD related small pathological features; Secondly, replace the dynamic full dimensional convolution module to optimize the multi-level extraction of image features; Finally, the coupling head strategy is adopted to improve the execution efficiency of classification and localization tasks separately.

RESULTS

We validated the effectiveness of the proposed method using a dataset of 582 MRI images from 108 patients. The results show that the proposed method achieves 0.961, 0.974, and 0.986 in Precision, Recall, and mAP, respectively, and the experimental results are superior to other algorithms.

CONSLUSION

The improved model has achieved high accuracy and detection accuracy, and can accurately detect and recognize complex Parkinson's MRI images.

SIGNIFICANCE

This algorithm has shown good performance in the early diagnosis of Parkinson's disease and can provide clinical assistance for doctors in early diagnosis. It compensates for the limitations of traditional methods.

A density map-based method for counting wheat ears

Introduction

Field wheat ear counting is an important step in wheat yield estimation, and how to solve the problem of rapid and effective wheat ear counting in a field environment to ensure the stability of food supply and provide more reliable data support for agricultural management and policy making is a key concern in the current agricultural field.

Methods

There are still some bottlenecks and challenges in solving the dense wheat counting problem with the currently available methods. To address these issues, we propose a new method based on the YOLACT framework that aims to improve the accuracy and efficiency of dense wheat counting. Replacing the pooling layer in the CBAM module with a GeM pooling layer, and then introducing the density map into the FPN, these improvements together make our method better able to cope with the challenges in dense scenarios.

Results

Experiments show our model improves wheat ear counting performance in complex backgrounds. The improved attention mechanism reduces the RMSE from 1.75 to 1.57. Based on the improved CBAM, the R2 increases from 0.9615 to 0.9798 through pixel-level density estimation, the density map mechanism accurately discerns overlapping count targets, which can provide more granular information.

Discussion

The findings demonstrate the practical potential of our framework for intelligent agriculture applications.

Maize plant detection using UAV-based RGB imaging and YOLOv5

In recent years, computer vision (CV) has made enormous progress and is providing great possibilities in analyzing images for object detection, especially with the application of machine learning (ML). Unmanned Aerial Vehicle (UAV) based high-resolution images allow to apply CV and ML methods for the detection of plants or their organs of interest. Thus, this study presents a practical workflow based on the You Only Look Once version 5 (YOLOv5) and UAV images to detect maize plants for counting their numbers in contrasting development stages, including the application of a semi-auto-labeling method based on the Segment Anything Model (SAM) to reduce the burden of labeling. Results showed that the trained model achieved a mean average precision (mAP@0.5) of 0.828 and 0.863 for the 3-leaf stage and 7-leaf stage, respectively. YOLOv5 achieved the best performance under the conditions of overgrown weeds, leaf occlusion, and blurry images, suggesting that YOLOv5 plays a practical role in obtaining excellent performance under realistic field conditions. Furthermore, introducing image-rotation augmentation and low noise weight enhanced model accuracy, with an increase of 0.024 and 0.016 mAP@0.5, respectively, compared to the original model of the 3-leaf stage. This work provides a practical reference for applying lightweight ML and deep learning methods to UAV images for automated object detection and characterization of plant growth under realistic environments.

A lightweight network for improving wheat ears detection and counting based on YOLOv5s

Introduction

Recognizing wheat ears plays a crucial role in predicting wheat yield. Employing deep learning methods for wheat ears identification is the mainstream method in current research and applications. However, such methods still face challenges, such as high computational parameter volume, large model weights, and slow processing speeds, making it difficult to apply them for real-time identification tasks on limited hardware resources in the wheat field. Therefore, exploring lightweight wheat ears detection methods for real-time recognition holds significant importance.

Methods

This study proposes a lightweight method for detecting and counting wheat ears based on YOLOv5s. It utilizes the ShuffleNetV2 lightweight convolutional neural network to optimize the YOLOv5s model by reducing the number of parameters and simplifying the complexity of the calculation processes. In addition, a lightweight upsampling operator content-aware reassembly of features is introduced in the feature pyramid structure to eliminate the impact of the lightweight process on the model detection performance. This approach aims to improve the spatial resolution of the feature images, enhance the effectiveness of the perceptual field, and reduce information loss. Finally, by introducing the dynamic target detection head, the shape of the detection head and the feature extraction strategy can be dynamically adjusted, and the detection accuracy can be improved when encountering wheat ears with large-scale changes, diverse shapes, or significant orientation variations.

Results and discussion

This study uses the global wheat head detection dataset and incorporates the local experimental dataset to improve the robustness and generalization of the proposed model. The weight, FLOPs and mAP of this model are 2.9 MB, 2.5 * 109 and 94.8%, respectively. The linear fitting determination coefficients R2 for the model test result and actual value of global wheat head detection dataset and local experimental Site are 0.94 and 0.97, respectively. The improved lightweight model can better meet the requirements of precision wheat ears counting and play an important role in embedded systems, mobile devices, or other hardware systems with limited computing resources.

A Multiscale Point-Supervised Network for Counting Maize Tassels in the Wild

Accurate counting of maize tassels is essential for monitoring crop growth and estimating crop yield. Recently, deep-learning-based object detection methods have been used for this purpose, where plant counts are estimated from the number of bounding boxes detected. However, these methods suffer from 2 issues: (a) The scales of maize tassels vary because of image capture from varying distances and crop growth stage; and (b) tassel areas tend to be affected by occlusions or complex backgrounds, making the detection inefficient. In this paper, we propose a multiscale lite attention enhancement network (MLAENet) that uses only point-level annotations (i.e., objects labeled with points) to count maize tassels in the wild. Specifically, the proposed method includes a new multicolumn lite feature extraction module that generates a scale-dependent density map by exploiting multiple dilated convolutions with different rates, capturing rich contextual information at different scales more effectively. In addition, a multifeature enhancement module that integrates an attention strategy is proposed to enable the model to distinguish between tassel areas and their complex backgrounds. Finally, a new up-sampling module, UP-Block, is designed to improve the quality of the estimated density map by automatically suppressing the gridding effect during the up-sampling process. Extensive experiments on 2 publicly available tassel-counting datasets, maize tassels counting and maize tassels counting from unmanned aerial vehicle, demonstrate that the proposed MLAENet achieves marked advantages in counting accuracy and inference speed compared to state-of-the-art methods. The model is publicly available at https://github.com/ShiratsuyuShigure/MLAENet-pytorch/tree/main.

Enhanced YOLOv5 Object Detection Algorithm for Accurate Detection of Adult Rhynchophorus ferrugineus

The red palm weevil (RPW, Rhynchophorus ferrugineus) is an invasive and highly destructive pest that poses a serious threat to palm plants. To improve the efficiency of adult RPWs' management, an enhanced YOLOv5 object detection algorithm based on an attention mechanism is proposed in this paper. Firstly, the detection capabilities for small targets are enhanced by adding a convolutional layer to the backbone network of YOLOv5 and forming a quadruple down-sampling layer by splicing and down-sampling the convolutional layers. Secondly, the Squeeze-and-Excitation (SE) attention mechanism and Convolutional Block Attention Module (CBAM) attention mechanism are inserted directly before the SPPF structure to improve the feature extraction capability of the model for targets. Then, 2600 images of RPWs in different scenes and forms are collected and organized for data support. These images are divided into a training set, validation set and test set following a ratio of 7:2:1. Finally, an experiment is conducted, demonstrating that the enhanced YOLOv5 algorithm achieves an average precision of 90.1% (mAP@0.5) and a precision of 93.8% (P), which is a significant improvement compared with related models. In conclusion, the enhanced model brings a higher detection accuracy and real-time performance to the RPW-controlled pest pre-detection system, which helps us to take timely preventive and control measures to avoid serious pest infestation. It also provides scalability for other pest pre-detection systems; with the corresponding dataset and training, the algorithm can be adapted to the detection tasks of other pests, which in turn brings a wider range of applications in the field of monitoring and control of agricultural pests.

Automatic rape flower cluster counting method based on low-cost labelling and UAV-RGB images

BACKGROUND

The flowering period is a critical time for the growth of rape plants. Counting rape flower clusters can help farmers to predict the yield information of the corresponding rape fields. However, counting in-field is a time-consuming and labor-intensive task. To address this, we explored a deep learning counting method based on unmanned aircraft vehicle (UAV). The proposed method developed the in-field counting of rape flower clusters as a density estimation problem. It is different from the object detection method of counting the bounding boxes. The crucial step of the density map estimation using deep learning is to train a deep neural network that maps from an input image to the corresponding annotated density map.

RESULTS

We explored a rape flower cluster counting network series: RapeNet and RapeNet+. A rectangular box labeling-based rape flower clusters dataset (RFRB) and a centroid labeling-based rape flower clusters dataset (RFCP) were used for network model training. To verify the performance of RapeNet series, the paper compares the counting result with the real values of manual annotation. The average accuracy (Acc), relative root mean square error (rrMSE) and [Formula: see text] of the metrics are up to 0.9062, 12.03 and 0.9635 on the dataset RFRB, and 0.9538, 5.61 and 0.9826 on the dataset RFCP, respectively. The resolution has little influence for the proposed model. In addition, the visualization results have some interpretability.

CONCLUSIONS

Extensive experimental results demonstrate that the RapeNet series outperforms other state-of-the-art counting approaches. The proposed method provides an important technical support for the crop counting statistics of rape flower clusters in field.