The segmentation of airborne laser scanning (ALS) point clouds remains a challenge in remote sensing and photogrammetry. Deep learning methods, such as KPCONV, have proven effective on various datasets. However, the rigid convolutional kernel strategy of KPCONV limits its potential use for 3D object segmentation due to its uniform approach. To address this issue, we propose an Integrated Point Convolution (IPCONV) based on KPCONV, which utilizes two different convolution kernel point generation strategies, one cylindrical and one a spherical cone, for more efficient learning of point cloud data features. We propose a customizable Multi-Shape Neighborhood System (MSNS) to balance the relationship between these convolution kernel point generations. Experiments on the ISPRS benchmark dataset, LASDU dataset, and DFC2019 dataset demonstrate the validity of our method.

This paper presents an automatic building extraction method which utilizes a photogrammetric digital surface model (DSM) and digital orthophoto map (DOM) with the help of historical digital line graphic (DLG) data. To reduce the need for manual labeling, the initial labels were automatically obtained from historical DLGs. Nonetheless, a proportion of these labels are incorrect due to changes (e.g., new constructions, demolished buildings). To select clean samples, an iterative method using random forest (RF) classifier was proposed in order to remove some possible incorrect labels. To get effective features, deep features extracted from normalized DSM (nDSM) and DOM using the pre-trained fully convolutional networks (FCN) were combined. To control the computation cost and alleviate the burden of redundancy, the principal component analysis (PCA) algorithm was applied to reduce the feature dimensions. Three data sets in two areas were employed with evaluation in two aspects. In these data sets, three DLGs with 15%, 65%, and 25% of noise were applied. The results demonstrate the proposed method could effectively select clean samples, and maintain acceptable quality of extracted results in both pixel-based and object-based evaluations.

Deep Neural Network (DNN)-based point cloud semantic segmentation has presented significant breakthrough using large-scale labeled aerial laser point cloud datasets. However, annotating such large-scaled point clouds is time-consuming. Self-Supervised Learning (SSL) is a promising approach to this problem by pre-training a DNN model utilizing unlabeled samples followed by a fine-tuned downstream task involving very limited labels. The traditional contrastive learning for point clouds selects the hardest negative samples by solely relying on the distance between the embedded features derived from the learning process, potentially evolving some negative samples from the same classes to reduce the contrastive learning effectiveness. This work proposes a hard-negative sample-aware self-supervised contrastive learning algorithm to pre-train the model for semantic segmentation. We designed a k-means clustering-based Absolute Positive And Negative samples (AbsPAN) strategy to filter the possible false-negative samples. Experiments on two typical ALS benchmark datasets demonstrate that the proposed method is more appealing than supervised training schemes without pre-training. Especially when the labels are severely inadequate (10% of the ISPRS training set), the results obtained by the proposed HAVANA method still exceed 94% of the supervised paradigm performance with full training set.

Detecting changes between the existing building basemaps and newly acquired high spatial resolution remotely sensed (HRS) images is a time-consuming task. This is mainly because of the data labeling and poor performance of hand-crafted features. In this paper, for efficient feature extraction, we propose a fully convolutional feature extractor that is reconstructed from the deep convolutional neural network (DCNN) and pre-trained on the Pascal VOC dataset. Our proposed method extract pixel-wise features, and choose salient features based on a random forest (RF) algorithm using the existing basemaps. A data cleaning method through cross-validation and label-uncertainty estimation is also proposed to select potential correct labels and use them for training an RF classifier to extract the building from new HRS images. The pixel-wise initial classification results are refined based on a superpixel-based graph cuts algorithm and compared to the existing building basemaps to obtain the change map. Experiments with two simulated and three real datasets confirm the effectiveness of our proposed method and indicate high accuracy and low false alarm rate.

When using digital images to obtain the displacement of a structure, noise is always a factor that greatly affects the accuracy of the displacement results. To overcome this problem, a phase-correlation improvement method, based on phase filtering by phase standard deviation (PSD), is proposed. We use PSD to remove the severe noise interference area in the cross-power spectrum matrix, then the singular value decomposition (SVD) is used to convert the two-dimensional phase to one-dimensional to reduce the difficulty of phase unwrapping. Finally, straight-line fitting is used to obtain the high-precision displacement results. Experiments were carried out using both simulated data and real data. For the simulated data, the absolute accuracy of the proposed method is around 0.3 pixels, while, for the real data, the absolute accuracy of the proposed method is 0.018 mm. The experimental results show that the method can effectively resist noise interference and still obtain accurate displacement estimation results in the case of severe noise.