Pine wilt is a devastating disease that typically kills affected pine trees within a few months. In this paper, we confront the problem of detecting pine wilt disease. In the image samples that have been used for pine wilt disease detection, there is high ambiguity due to poor image resolution and the presence of “disease-like” objects. We therefore created a new dataset using large-sized orthophotographs collected from 32 cities, 167 regions, and 6121 pine wilt disease hotspots in South Korea. In our system, pine wilt disease was detected in two stages: n the first stage, the disease and hard negative samples were collected using a convolutional neural network. Because the diseased areas varied in size and color, and as the disease manifests differently from the early stage to the late stage, hard negative samples were further categorized into six different classes to simplify the complexity of the dataset. Then, in the second stage, we used an object detection model to localize the disease and “disease-like” hard negative samples. We used several image augmentation methods to boost system performance and avoid overfitting. The test process was divided into two phases: a patch-based test and a real-world test. During the patch-based test, we used the test-time augmentation method to obtain the average prediction of our system across multiple augmented samples of data, and the prediction results showed a mean average precision of 89.44% in five-fold cross validation, thus representing an increase of around 5% over the alternative system. In the real-world test, we collected 10 orthophotographs in various resolutions and areas, and our system successfully detected 711 out of 730 potential disease spots.
22 January 2024 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Multispectral image pansharpening is a significant technology for remote sensing image analysis, aiming to restore a high-resolution multispectral (HRMS) image by merging a high-resolution panchromatic (PAN) image with a low-resolution multispectral (MS) image. While the convolutional neural networks (CNN) have garnered considerable attention for their exceptional fusion capabilities in recent years, existing CNN-based methods cannot effectively integrate spectral-spatial information. In this paper, we introduce a novel multispectral pansharpening framework that integrates spectral and spatial networks in a bilateral pyramid form, allowing for the extraction of hierarchical spectral-spatial information. The proposed reduced residual dense (RRD) module serves as the fundamental building block of the spatial network. The RRD module gradually reduces the dimension of feature maps and employs the concatenation of RRD with global residual learning for comprehensive feature representation. In the spectral network, we present a cooperative attention fusion (CAF) module to further enhance the correlation between spectral and spatial features. Through extensive experiments conducted on benchmarking simulated and real datasets, our proposed framework consistently outperforms state-of-the-art methods, demonstrating its effectiveness in multispectral image pansharpening applications.
Pine wilt disease (PWD) is responsible for significant damage to East Asia’s pine forests, including those in Korea, Japan, and China. Preventing the spread of wilt disease requires early detection and removal of damaged trees. This paper proposes a method of detecting disease-damaged pines using ortho-images corrected from 5-band multi-spectral images captured by unmanned aviation vehicles. The proposed method relies on a ResNet18 backbone network connected to a modified DenseNet module, classifies the 5-band multispectral (RGB, NIR, Red_Edge) ortho-image patches, and visualizes the results as a heat map. The patch-based classifier was retrained with hard negative examples, after which it achieved 98.66% accuracy, an improvement over the 96.0% accuracy associated with the same method applied to RGB images. The resulting heat map reflects the approximate distribution, and movement of the disease. Disease locations are also predicted by local maximums in the heat map. When the distance between a ground truth and the predicted location is less than visible distance, e.g. about 5m, it is counted as a correct detection. The proposed detection which consists of heat map generation followed by localization achieves Recall of 93.39%, Precision of 88.26%, and F1-score of 90.75%.
This study presents an offline mobile diagnosis system for citrus pests and diseases by compression convolutional neural network. Recently, with the growth of labelled data, the deep neural network incites the revolutionary change with a quantum leap in various fields. Benefiting from the backpropagation method, the proper network structure can automatically extract high‐level representations and find corresponding labels. The authors made use of the advantages of the deep neural network to design an android application, which can be installed in any stand‐alone devices to instantaneously identify the citrus pests and diseases. The proposed diagnosis system has three characteristics: low cost, low latency and high accuracy. These characteristics contribute to make the professional offline prediction for avoiding further economic loss caused by disease spreading. In order to validate the proposed system, the authors conducted thorough evaluations on two data sets, ‘citrus pests and diseases’, CIFAR, which show the superiority of the proposed approach in terms of the accuracy and the number of model parameters.
14 September 2022 in Circuits, Systems, and Signal Processing
Music source separation has traditionally followed the encoder-decoder paradigm (e.g., hourglass, U-Net, DeconvNet, SegNet) to isolate individual music components from mixtures. Such networks, however, result in a loss of location-sensitivity, as low-resolution representation drops the useful harmonic patterns over the temporal dimension. We overcame this problem by performing singing voice separation using a high-resolution representation learning (HRNet) system coupled with a long short-term memory (LSTM) module to retain high-resolution feature map and capture the temporal behavior of the acoustic signal. We called this joint combination of HRNet and LSTM as HR-LSTM. The predicted spectrograms produced by this system are close to ground truth and successfully separate music sources, achieving results superior to those realized by past methods. The proposed network was tested using four datasets (DSD100, MIR-1K, Korean Pansori, and Nepal Idol singing voice). Our experiments confirmed that the proposed HR-LSTM outperforms state-of-the-art networks at singing voice separation when the DSD100 dataset is used, performs comparably to alternative methods when the MIR-1K dataset is used, and separates the voice and accompaniment components well when the Pansori and NISVS datasets are used. In addition to proposing and validating our network, we also developed and shared our Nepal Idol dataset.
Pansori, a traditional Korean form of musical storytelling, is characterized by performances involving a vocalist and a drummer. It is well-known for the singer’s expressive narrative (aniri) and delicate gesture with fan in hand. The classical Pansori repertoires mostly tell love, satire, and humor, as well as some social lessons. These performances, which can extend from three to five hours, necessitate that the vocalist adheres to precise rhythmic structures. The distinctive rhythms of Pansori are crucial for conveying both the narrative and musical expression effectively. This paper explores the challenge of open-set recognition, aiming to efficiently identify unknown Pansori rhythm patterns while applying the methodology to diverse acoustic datasets, such as sound events and genres. We propose a lightweight deep learning-based encoder–decoder segmentation model, which employs a 2-D log-Mel spectrogram as input for the encoder and produces a frame-based 1-D decision along the temporal axis. This segmentation approach, processing 2-D inputs to classify frame-wise rhythm patterns, proves effective in detecting unknown patterns within time-varying sound streams encountered in daily life. Throughout the training phase, both center and supervised contrastive losses, along with cross-entropy loss, are minimized. This strategy aimed to create a compact cluster structure within the feature space for known classes, thereby facilitating the recognition of unknown rhythm patterns by allocating ample space for their placement within the embedded feature space. Comprehensive experiments utilizing various datasets—including Pansori rhythm patterns (91.8%), synthetic datasets of instrument sounds (95.1%), music genres (76.9%), and sound datasets from DCASE challenges (73.0%)—demonstrate the efficacy of our proposed method to detect unknown events, as evidenced by the AUROC metrics.