Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Qiufeng Wu has a total of 56 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Droplet-YOLO: Rice Guttation Droplets Detection Based on YOLOv8 and Multi-Instance Learning

    by Chuanhui Gong, Qiufeng Wu

    Guttation plays an important role in increasing rice yield, preventing crop diseases and improving soil fertility, and is an important index to measure the water status in the field. However, the real-time detection of droplets generated by guttation is a difficult task. Droplets are small and dense targets. In the deep learning model, the detection of small objects in high-resolution images is a key problem to be solved. A Droplet-YOLO method is proposed, which combines the YOLOv8 model and multi-instance learning. The original image is divided into multiple sub-images using a sliding window and detected by YOLOv8. The detection of each sub-image is carried out by an independent repeated test, so multi-instance learning ensures 99.99% probability of detecting droplets. By comparing various models, Droplet-YOLO performs best in terms of precision, recall, and mean precision (mAP). In addition, through the hyper-parameter adjustment experiment, the configuration with a batch size of 32 and epoch of 200 was finally selected. The accuracy and recall rate of the model reached 96.8% and 95.7%, respectively, and the mAP reached 99.0%. Experiments show that this method has significant advantages for small object detection in high-resolution images, and the proposed Droplet-YOLO model is superior to the most advanced methods.

  • Open Access

    A DoveNet-Based Method for Plant Disease Image Generation

    by Xinyue Sun, Xiangyan Meng, Qiufeng Wu

    Image generation of plant disease in the natural environment has always been a challenging task. Traditional methods applied in the image generation of plant disease are without sufficient diversity and detailed lesions. Thus, this paper applies an image harmonization method to generate diverse combinations of disease images by integrating different backgrounds and target regions to enhance diversity. To construct the dataset, we captured real disease images of soybean and rice in natural environments. Next, the Squeeze-and-Excitation (SE) attention mechanism was integrated into the domain verification network (DoveNet), together with a mask guide generator, to focus more attention on lesions. Two discriminators worked together to capture global and local features, ensuring the preservation of critical contextual information. Finally, the improved DoveNet achieved a MSE of 43.77, a PSNR of 33.02, and an SSIM of 0.9806, showing a reduction of 3.61 in the MSE, an increase of 0.50 in the PSNR, and a 2.49% improvement in the SSIM compared with the original DoveNet. Meanwhile, through visual Turing tests we confirmed that images generated using the improved DoveNet were of much better quality and more convincing.

  • Open Access

    Improved Unsupervised Domain Generalization by Leaf-Bridge for Soybean Leaf Disease Identification

    by Jiaqi Cui, Jinpeng Gu, Qiufeng Wu

    Soybean diseases are the main factors causing serious yield reduction in soybeans. Therefore, an automatic identification method for soybean leaf diseases is an urgent need to develop. However, most soybean leaf disease recognition models have problems such as weak generalization ability and insufficient labeled soybean images. To overcome these problems, this paper proposes an improved method for unsupervised domain generalization (UDG) named Leaf-Bridge Across Domains (LBADs), which is dedicated to cross-domain soybean leaf disease image recognition. Our method builds a semantic-preserving image-to-image mapping between each training domain and an auxiliary leaf-bridge domain to achieve cross-domain semantic alignment of soybean disease images in the shared feature space through fully unsupervised self-supervised contrastive learning. Firstly, LBAD is performed on both labeled source and target domain data. To bridge the gap between these two domains, it constructs a cross-domain aided leaf-bridge domain within a contrastive self-supervised model, mapping images from different domains into a shared space to achieve semantic alignment. After the fully unsupervised pre-training of the feature backbone, we use 1%, 5%, and 10% of the labeled source domain images, respectively, to fine-tune the linear classifier only (the pre-trained backbone weights are completely frozen) and complete the classification of unlabeled target domain images. The highest accuracy of the three-label fraction can reach 83.56%, 89.22%, and 87.2%, respectively, higher than that of the model without LBAD. The final identification result is also higher than the accuracy of the model trained without LBAD. Experimental results showed that LBAD can improve the generalization ability of the model and obtain a more accurate soybean leaf disease recognition model.

  • Open Access

    Fruit Detection Methods Based on Deep Learning in Agricultural Planting: A Systematic Literature Review

    by Xinyu Gong, Qiufeng Wu

    With the continued advancement of agricultural intelligence, automation, and mechanization, artificial intelligence-driven fruit detection technology has developed rapidly. As an important task in agricultural computer vision, fruit target detection in real-world planting environments presents numerous technical challenges. This paper provides a systematic review of recent breakthroughs and representative studies in this field. Based on a comprehensive analysis of existing research, we classify deep learning–based fruit detection models into four application scenarios: few-shot detection (addressing limited data availability and high annotation costs), complex scene detection (resolving issues arising from object occlusion, overlapping, and variable illumination), small-target detection (improving performance on low-resolution and densely clustered objects), and real-time detection (designing lightweight algorithms for faster inference). The paper summarizes innovative technical approaches and evaluates detection performance across these scenarios, highlighting recent advancements in fruit detection for agricultural production and offering valuable insights for further technological innovation.


  • Open Access

    Fourier Domain Adaptation for the Identification of Grape Leaf Diseases

    by Jing Wang, Qiufeng Wu, Tianci Liu, Yuqi Wang, Pengxian Li, Tianhao Yuan, Ziyang Ji

    With the application of computer vision in the field of agricultural disease recognition, the convolutional neural network is widely used in grape leaf disease recognition and has achieved remarkable results. However, most of the grape leaf disease recognition models have the problem of weak generalization ability. In order to overcome this challenge, this paper proposes an image identification method for grape leaf diseases in different domains based on Fourier domain adaptation. Firstly, Fourier domain adaptation is performed on the labeled source domain data and the unlabeled target domain data. To decrease the gap in distribution between the source domain data and the target domain data, the low-frequency spectrum of the source domain data and the target domain data is swapped. Then, three convolutional neural networks (AlexNet, VGG13, and ResNet101) were used to train the images after style changes and the unlabeled target domain images were classified. The highest accuracy of the three networks can reach 94.6%, 96.7%, and 91.8%, respectively, higher than that of the model without Fourier transform image training. In order to reduce the impact of randomness, when selecting the transformed image, we propose using farthest point sampling to select the image with low feature correlation for the Fourier transform. The final identification result is also higher than the accuracy of the network model trained without transformation. Experimental results showed that Fourier domain adaptation can improve the generalization ability of the model and obtain a more accurate grape leaf disease recognition model.


  • Open Access

    Application of Improved Lightweight Network and Choquet Fuzzy Ensemble Technology for Soybean Disease Identification

    by Yan Hang, Xiangyan Meng, Qiufeng Wu

    The identification of soybean disease images in natural scenes has been a challenging task due to their complex backgrounds and diverse spot patterns. Traditional single convolutional neural network (CNN) for soybean disease image recognition often cannot have both high identification accuracy and strong generalization ability. Therefore, this paper focuses on the classification of soybean leaf diseases using improved lightweight networks for transfer learning, and improves the identification accuracy and precision by introducing Choquet fuzzy ensemble strategy. First, the convolutional long short-term memory (ConvLSTM) layer and the squeeze and excitation (SE) block are introduced into the four original lightweight models (Xception, MobileNetV2, NASNetMobile, MobileNet) to improve the network’s ability to grasp image features, and then the classification confidence scores obtained from the improved lightweight networks are fed into the fuzzy iensemble network to complete the aggregation of the final results. In order to improve the performance of the model and enrich the distribution of samples in the high-dimensional feature space, this paper converts soybean healthy leaf images to diseased leaf images using an unsupervised image translation method based on Cycle-Consistent Adversarial Networks (CycleGAN). The results show that the improved lightweight model has higher recognition accuracy than the original network. The proposed fuzzy ensemble model obtains 94.27% recognition accuracy and an average F1-score of 94% in the soybean leaf disease classification task, which is better than a single model and other ensemble methods. It has a good application prospect and initially meets the production requirements of soybean disease identification.


  • Open Access

    Apple leaf disease identification via improved CycleGAN and Convolutional Neural Network

    by Yiping Chen, Jinchao Pan, Qiufeng Wu

    The identification of apple leaf diseases is crucial to reduce yield reduction and timely take disease control measures. Employing machine learning-based methods, such as deep learning for accurate identification of multiple apple leaf diseases is challenging because of the limited availability of samples for supervised training and the serious class imbalance. Hence, this paper proposes an accurate deep learning-based pipeline to solve the problem of limited data sets on farms and reduce the partiality due to serious class imbalance. Firstly, this paper proposes the improved cycle-consistent adversarial networks (CycleGAN) to generate synthetic samples to improve the learning of data distribution and solve the problems of small data sets and class imbalance. On the basis of CycleGAN, two discriminators are introduced: one to judge whether the image is true or false, and one to judge whether the leaf image has disease. Specifically, healthy leaves are converted into disease-carrying leaves using two models, health-to-scab and health-to-rust to balance apple leaf disease datasets. Secondly, ResNet is trained as a baseline convolutional neural network classifier to classify apple leaf diseases. In experimental results, this paper carries out some experiments on evaluation of quality of the generated images by the improved CycleGAN and the performance of ResNet in terms of datasets and metrics. First of all, through qualitative observation of generated images and quantitative metrics, such as GAN-train and GAN-test, the generated diseases images from healthy image by the improved CycleGAN are superior to state of the art. Second, some experiments were carried out to verify the performance of the classification model, such as comparing five mainstream classification models and comparing them with traditional data augmentation methods. The results show that ResNet has the highest recognition accuracy on the test set, reaching 97.78%, and the classification accuracy is significantly improved by the generated synthetic samples (+14.7%). Finally, the experiment result of t-Distributed Stochastic Neighbor Embedding (t-SNE) visually confirmed that the images generated by improved CycleGAN have much better quality and are more convincing. Beyond that, a Visual Turing Test with three botanists showed that the generated images are indistinguishable from real apple leaf images.