The early growth process of wheat is vulnerable to various factors, and poor growth leads to vacancies in the planting row. Therefore, the wheat images captured by unmanned aerial vehicles (UAV) are essential for monitoring the growth of wheat and preventing diseases and insect pests. This paper uses wheat images captured by UAV as a dataset, and propose a novel residual-capsule network with threshold convolution (RCTC) for segmentation of wheat plantation rows. The network is achieved by replacing the AveragePooling of the improved ResNet34 with Capsule. Since the capsule network represents the features by vectors, it can explain the direction of features and the relative positions between features. Therefore, deeper feature information can be extracted. In addition, to reduce redundant features and enhance effective features, a new threshold convolution is also proposed. Experiments on the wheat field dataset show that our proposed algorithm can effectively segment the wheat plantation rows images collected by UAV, and is superior to some existing well-known algorithms, and can provide scientific support and reference for the decision-making process of smart agriculture.

Unsupervised polarimetric synthetic aperture radar (PolSAR) image classification is an important task in PolSAR automatic image analysis and interpretation. Generally, a group of features is insufficient to effectively classify PolSAR images, especially in multiple terrain scenarios. Therefore, multiple features need to be extracted for PolSAR image classification. However, how to combine and integrate these features effectively to fully utilize each feature's information and discriminability need to be determined. Such integrated work has traditionally received little attention. In this paper, a novel unsupervised classification framework for PolSAR images is proposed. First, a PolSAR image is oversegmented via a fast superpixel segmentation method. Second, five feature vectors are extracted from PolSAR images via superpixels, resulting in five corresponding similarity matrices that are constructed by using Gaussian kernels. Third, consensus similarity network fusion (CSNF), originally proposed and widely used for biomedical sciences, is employed to combine and integrate the five similarity matrices to obtain a fused similarity matrix. Fourth, spectral clustering method, based on the fused similarity matrix, is used to cluster the PolSAR image. Finally, a novel classification postprocessing procedure is presented and exploited to smooth the initial clusters and correct some misclassified pixels. The extensive experimental results conducted on one simulated and two real-world PolSAR images demonstrate the feasibility and superiority of the proposed method compared with five other state-of-the-art classification approaches. INDEX TERMS Polarimetric synthetic aperture radar (PolSAR) images, unsupervised classification, superpixels segmentation, consensus similarity network fusion (CSNF), spectral clustering.

No abstract

Clustering-based methods of polarimetric synthetic aperture radar (PolSAR) image superpixel generation are popular due to their feasibility and parameter controllability. However, these methods pay more attention to improving boundary adherence and are usually time-consuming to generate satisfactory superpixels. To address this issue, a novel cross-iteration strategy is proposed to integrate various advantages of different distances with higher computational efficiency for the first time. Therefore, the revised Wishart distance (RWD), which has better boundary adherence but is time-consuming, is first integrated with the geodesic distance (GD), which has higher efficiency and more regular shape, to form a comprehensive similarity measure via the cross-iteration strategy. This similarity measure is then utilized alternately in the local clustering process according to the difference between two consecutive ratios of the current number of unstable pixels to the total number of unstable pixels, to achieve a lower computational burden and competitive accuracy for superpixel generation. Furthermore, hexagonal initialization is adopted to further reduce the complexity of searching pixels for relabelling in the local regions. Extensive experiments conducted on the AIRSAR, RADARSAT-2 and simulated data sets demonstrate that the proposed method exhibits higher computational efficiency and a more regular shape, resulting in a smooth representation of land cover in homogeneous regions and better-preserved details in heterogeneous regions.

No abstract