Atmospheric fine particles (${\mathrm{PM}}_{2.5}$) have been found to be harmful to the environment and human health. Recently, remote sensing technology and machine learning models have been used to monitor ${\mathrm{PM}}_{2.5}$ concentrations. Partial dependence plots (PDP) were used to explore the meteorology mechanisms between predictor variables and ${\mathrm{PM}}_{2.5}$ concentration in the “black box” models. However, there are two key shortcomings in the original PDP. (1) it calculates the marginal effect of feature(s) on the predicted outcome of a machine learning model, therefore some local effects might be hidden. (2) it requires that the feature(s) for which the partial dependence is computed are not correlated with other features, otherwise the estimated feature effect has a great bias. In this study, the original PDP’s shortcomings were analyzed. Results show the contradictory correlation between the temperature and the ${\mathrm{PM}}_{2.5}$ concentration that can be given by the original PDP. Furthermore, the spatiotemporal heterogeneity of ${\mathrm{PM}}_{2.5}$-AOD relationship cannot be displayed well by the original PDP. The drawbacks of the original PDP make it unsuitable for exploring large-area feature effects. To resolve the above issue, multi-way PDP is recommended, which can characterize how the ${\mathrm{PM}}_{2.5}$ concentrations changed with the temporal and spatial variations of major meteorological factors in China.

The dropout and data augmentation techniques are widely used to prevent a convolutional neural network (CNN) from overfitting. However, the dropout technique does not work well when applied to the input channels of neural networks, and data augmentation is usually employed along the image plane. In this letter, we present DropBand, which is a simple and effective method of promoting the classification accuracy of CNNs for very-high-resolution remote sensing image scenes. In DropBand, more training samples are generated by dropping certain spectral bands out of original images. Furthermore, all samples with the same set of spectral bands are collected together to train a base CNN. The final prediction for a test sample is represented by the combination of outputs of all base CNNs. The experimental results for three publicly available data sets, i.e., the SAT-4, SAT-6, and UC-Merced image data sets, show that DropBand can significantly improve the classification accuracy of a CNN.

Modern convolutional neural networks (CNNs) are often trained on pre-set data sets with a fixed size. As for the large-scale applications of satellite images, for example, global or regional mappings, these images are collected incrementally by multiple stages in general. In other words, the sizes of training datasets might be increased for the tasks of mapping rather than be fixed beforehand. In this paper, we present a novel algorithm, called GeoBoost, for the incremental-learning tasks of semantic segmentation via convolutional neural networks. Specifically, the GeoBoost algorithm is trained in an end-to-end manner on the newly available data, and it does not decrease the performance of previously trained models. The effectiveness of the GeoBoost algorithm is verified on the large-scale data set of DREAM-B. This method avoids the need for training on the enlarged data set from scratch and would become more effective along with more available data.

Satellite images are always partitioned into regular patches with smaller sizes and then individually fed into deep neural networks (DNNs) for semantic segmentation. The underlying assumption is that these images are independent of one another in terms of geographic spatial information. However, it is well known that many land-cover or land-use categories share common regional characteristics within a certain spatial scale. For example, the style of buildings may change from one city or country to another. In this paper, we explore some deep learning approaches integrated with geospatial hash codes to improve the semantic segmentation results of satellite images. Specifically, the geographic coordinates of satellite images are encoded into a string of binary codes using the geohash method. Then, the binary codes of the geographic coordinates are fed into the deep neural network using three different methods in order to enhance the semantic segmentation ability of the deep neural network for satellite images. Experiments on three datasets demonstrate the effectiveness of embedding geographic coordinates into the neural networks. Our method yields a significant improvement over previous methods that do not use geospatial information.

Stochastic gradient descent and other adaptive optimization methods have been proved effective for training deep neural networks. Within each epoch of these methods, the whole training set is involved to train the model. In general, large training data sets have data redundancy among their training samples. In this paper, we present an algorithm to reduce the training time consumption of CNN by dropping certain samples out, which is called the greedy DropSample. For the absence of certain training samples, the distribution of networks’ activations is biased during training. By correcting the mean and variance of batch-normalization layers, this issue is solved. Experimental results over several data sets demonstrate the efficiency of the proposed method. The results show that this method could decrease the training time of multilayer perceptrons (MLPs) and convolutional neural networks (CNNs) significantly. Despite the reduced number of training samples, the accuracies of networks are similar, or even better.