Farmland parcel-based crop classification using satellite data plays an important role in precision agriculture. In this study, a deep-learning-based time-series analysis method employing optical images and synthetic aperture radar (SAR) data is presented for crop classification for cloudy and rainy regions. Central to this method is the spatial-temporal incorporation of high-resolution optical images and multi-temporal SAR data and deep-learning-based time-series analysis. First, a precise farmland parcel map is delineated from high-resolution optical images. Second, pre-processed SAR intensity images are overlaid onto the parcel map to construct time series of crop growth for each parcel. Third, a deep-learning-based (using the long short-term memory, LSTM, network) classifier is employed to learn time-series features of crops and to classify parcels to produce a final classification map. The method was applied to two datasets of high-resolution ZY-3 images and multi-temporal Sentinel-1A SAR data to classify crop types in Hunan and Guizhou of China. The classification results, with an 5.0% improvement in overall accuracy compared to those of traditional methods, illustrate the effectiveness of the proposed framework for parcel-based crop classification for southern China. A further analysis of the relationship between crop calendars and change patterns of time-series intensity indicates that the LSTM model could learn and extract useful features for time-series crop classification.

The time series (TS) of the normalized difference vegetation index (NDVI) has been widely used to trace the temporal and spatial variability of terrestrial vegetation. However, many factors such as atmospheric noise and radiometric correction residuals conceal the actual variation in the land surface, and thus hamper the TS information extraction. To minimize the negative effects of these noise factors, we propose a new method to produce a synthetic gap-free NDVI TS from the original contaminated observation. First, the key temporal points are identified from the NDVI time profiles based on a generally used rule-based strategy, making the TS segmented into several adjacent segments. Then, the observed data points in each segment are fitted with a weighted double-logistic function. The proposed dynamic weight reassignment process effectively emphasizes cloud-free points and deemphasizes cloud-contaminated points. Finally, the proposed method is evaluated on more than 3,000 test points from three selected Sentinel-2 tiles, and is compared with the generally used Savitzky-Golay (S-G) and harmonic analysis of time series (HANTS) methods from qualitative and quantitative aspects. The results indicate that the proposed method has a higher capability of retaining cloud-free data points and identifying outliers than the others, and can generate a gap-free NDVI time profile derived from a medium-resolution satellite sensor.

Geo-parcel based crop identification plays an important role in precision agriculture. It meets the needs of refined farmland management. This study presents an improved identification procedure for geo-parcel based crop identification by combining fine-resolution images and multi-source medium-resolution images. GF-2 images with fine spatial resolution of 0.8 m provided agricultural farming plot boundaries, and GF-1 (16 m) and Landsat 8 OLI data were used to transform the geo-parcel based enhanced vegetation index (EVI) time-series. In this study, we propose a piecewise EVI time-series smoothing method to fit irregular time profiles, especially for crop rotation situations. Global EVI time-series were divided into several temporal segments, from which phenological metrics could be derived. This method was applied to Lixian, where crop rotation was the common practice of growing different types of crops, in the same plot, in sequenced seasons. After collection of phenological features and multi-temporal spectral information, Random Forest (RF) was performed to classify crop types, and the overall accuracy was 93.27%. Moreover, an analysis of feature significance showed that phenological features were of greater importance for distinguishing agricultural land cover compared to temporal spectral information. The identification results indicated that the integration of high spatial-temporal resolution imagery is promising for geo-parcel based crop identification and that the newly proposed smoothing method is effective.

Thin clouds in remote sensing images increase the radiometric distortion of land surfaces. The identification of pixels contaminated by thin clouds, known as the thin-cloud mask, is an important preprocessing procedure to guarantee the proper utilization of data. However, failure to effectively separate thin clouds and high-reflective land-cover features causes thin-cloud masks to remain a challenge. To overcome this problem, we developed a thin-cloud masking method for remote sensing images based on sparse dark pixel region detection. As a result of the effect of scattering, the path radiance is added to the radiance recorded by the sensor in the thin-cloud area, which causes the number of dark pixels in the thin-cloud area to be much less than that in the clear area. In this study, the area of a Thiessen polygon (a nonparametric measure) is used to evaluate the density of local dark pixels, and the region with the sparse dark pixel is selected as the thin-cloud candidate. Then, thin-cloud and clear areas are used as samples to train the background suppression haze thickness index (BSHTI) transform parameters, and convert the original multiband images into single-band images. Finally, an accurate thin-cloud mask is obtained for every buffered thin-cloud candidate, via the segmentation of the BSHTI band. Additionally, the multispectral images obtained by the Wide Field View (WFV), on board the Chinese GaoFen1, and the Operational Land Imager (OLI), on board the Landsat 8, are employed to evaluate the performance of the method. The results reveal that the proposed approach can obtain a thin-cloud mask with a high true-value ratio and detection ratio. Thin-cloud masks can satisfy various application demands.

Building change detection (BuCD) can offer fundamental data for applications such as urban planning and identifying illegally-built new buildings. With the development of deep neural network-based approaches, BuCD using high-spatial-resolution remote sensing images (RSIs) has significantly advanced. These deep neural network-based methods, nevertheless, typically demand a considerable number of computational resources. Additionally, the accuracy of these algorithms can be improved. Hence, LightCDNet, a lightweight Siamese neural network for BuCD, is introduced in this paper. Specifically, LightCDNet comprises three components: a Siamese encoder, a multi-temporal feature fusion module (MultiTFFM), and a decoder. In the Siamese encoder, MobileNetV2 is chosen as the feature extractor to decrease computational costs. Afterward, the multi-temporal features from dual branches are independently concatenated based on the layer level. Subsequently, multiscale features computed from higher levels are up-sampled and fused with the lower-level ones. In the decoder, deconvolutional layers are adopted to gradually recover the changed buildings. The proposed network LightCDNet was assessed using two public datasets: namely, the LEVIR BuCD dataset (LEVIRCD) and the WHU BuCD dataset (WHUCD). The F1 scores on the LEVIRCD and WHUCD datasets of LightCDNet were 89.6% and 91.5%, respectively. The results of the comparative experiments demonstrate that LightCDNet outperforms several state-of-the-art methods in accuracy and efficiency.

The U-Net-like model has been widely studied in the field of building extraction. However, most of these models are based on locally sensed Convolutional Neural Networks(CNNs) designed with symmetric structure and single feature processing, which cannot accurately identify buildings with different sizes, shapes, and colors in remote sensing images. To overcome these problems, we propose the asymmetric cascade fusion network(ACFN), based on the Vision Transformer(ViT), to design a novel asymmetric architecture to recognize buildings of different sizes and shapes by processing multi-granularity features by different means. First, the asymmetric architecture obtains multi-granularity features with global contextual information by embedding different types of attention in encoder-decoders of different sizes. This architecture can identify densely distributed and occluded buildings by semantic reasoning in remote sensing images with complex information. Second, we design a multi-branch weighted pyramid pooling module, which sets different branch weights to offset the background noise introduced in introducing global contextual information. Our ACFN significantly improves the Beijing buildings, ISPRS-Vaihingen, and LoveDA datasets.