Search

• The ISP method allows for gathering a large number of historical training samples. • The SRVI values of the same crop planted in neighboring years range from 0.91 and 1.2. • A distinct northward expansion trend is evident for all four major crops, in Heilongjiang Province, China. Mapping agricultural information such as cropping area and type is of great significance for land use and food security and a common method to retrieve such information is satellite imagery classification. The current image classification techniques require a statistically large number of training samples to obtain representative spectral signatures, ideally collected during the satellite overpass time/date. Although this prerequisite sounds simple and easy in principle, it is extremely challenging and sometimes impossible when classifying historical satellite images, as one simply cannot go back in time to collect training datasets or ground truthing. In this paper, we introduce an iterative and subsequent proximation (ISP) method to circumvent this problem for historical image classifications for historical crop mapping. This method assumes that the same crops grown two years apart have similar spectral properties within the same growing season, with limited variation. This allows the use of the crop information classified in any given year (n) to be used as training samples for the next year (n + 1) or the previous year (n-1). Iteratively repeating n-1, n-2... process leads to the mapping of historical cropping information without a priori ground truthing data. To demonstrate the ISP feasibility, we first used the historical Landsat time-series data to map four major stable crops (rice, maize, soybean, and wheat) in Hailun County, Heilongjiang Province, China, and then further expanded the ISP application to the entire Heilongjiang Province to map cropping areas from 1982 to 2020, using the classification and regression trees (CART) algorithm. The results were validated for those years when actual cropping measurement data were available and further verified with statistical data for other years. The results indicated that the proposed ISP method was appropriate for the historical mapping of four major crops, with a mapping accuracy of approximately 80 % when validated with field data, and correlation coefficients of 0.86, 0.85, 0.91, and 0.94, respectively when compared with historical statistical cropping data. The crop mapping results showed distinct trends of northward expansion in each of the four crops in Heilongjiang Province, which agreed well with previous studies. In conclusion, the ISP method is quick, easy, and convenient for historical crop mapping, which is important in understanding the agricultural production history. [ABSTRACT FROM AUTHOR]

• Sensor technology developments drove advances in biophysical estimation techniques. • CubeSat constellations will increase VHR data accessibility to various users. • Sentinel-2 red-edge bands present prospects for operational crop stress detection. • RF is prominently used in the literature despite new developments in algorithms. • Retrieval models developed in Mediterranean areas do not apply to semi-arid areas. This paper provides an overview of the recent developments in remote sensing technology and machine learning algorithms for estimating important biophysical and biochemical parameters for precision farming. The objectives are (i) to provide an overview of recent advances in remotely sensed retrieval of biophysical and biochemical parameters brought by the developments in sensor technologies and robust machine learning algorithms and (ii) to identify the sources of uncertainty in retrieving biophysical and biochemical parameters and implications for precision agriculture. The review revealed that developments in crop biophysical and biochemical parameters retrieval techniques were mainly driven by announcements and the availability of new sensors. Two ground-breaking events can be identified, i.e., the availability of Sentinel-2 and the SuperDove constellation. The two provide high temporal-high spatial resolution data relevant for site-specific management and super-spectral configuration, enabling retrieval of crop growth and health parameters. The free availability of Sentinel-2 triggered the testing of its spectral configurations and upscaling of retrieval approaches using simulated data from field spectrometers and airborne hyperspectral sensors. SuperDoves will likely reduce the cost of very high-resolution data while providing unprecedented capabilities for detailed, accurate and frequent characterisation of field variability. Studies showed that the red-edge bands and hybrid models coupling Radiative Transfer Model (RTM) and machine learning regression algorithms (MLRA) are promising for operational and accurate monitoring of stress-related crop parameters to aid time-sensitive agronomic decisions. However, such models were tested in Mediterranean climates and performed poorly in African semi-arid areas and China's temperate continental semi-humid monsoon climates. Therefore, locally-calibrated RTM models incorporating crop-type maps and other spatio-temporal constraints may reduce uncertainties when adapted to data-scarce regions. Generally, permanent experimental sites and a lack of systematic calibration data on various crops are some limiting factors to using remote sensing technologies for PA in Sub-Saharan Africa. Other complexities arise from farm configurations, such as small field sizes and mixed cropping practices. Therefore, future studies should develop generic, scalable and transferable models, especially within under-studied areas. [ABSTRACT FROM AUTHOR]

• The monitoring and estimation method of Winter Wheat Soil Moisture Based on multi-source sensing data and CNNR model is studied in this paper. • The proposed CNNR model has the best prediction accuracy. With the decrease of the SNR of radar echo, CNNR always has higher R2 and RPD, which indicates that the CNNR model can better predict the change trend close to the real value. Therefore, CNNR model has the best noise robustness among the three models. Rapid and accurate estimation of soil moisture content (SMC) is an important part of precision agriculture, and it is also one of the key problems to be solved in field real-time monitoring and precision irrigation. Most of the existing studies are limited to SMC monitoring of bare soil, which can be obtained by optical remote sensing (e.g., multispectral, hyperspectral) or Synthetic Aperture Radar (SAR). However, for the soil covered by vegetation, such as farmland, there are some theoretical defects with only one of the measuring methods. Meanwhile, in order to break through the limitations of low spatial and temporal resolutions of satellite remote sensing, it is of great significance to study SMC retrieval based on multi-source remote sensing data for the near earth UAV remote sensing systems. Based on this, this paper, taking the winter wheat planting area in Guanzhong plain of China as the research area, combines the advantages of ultra-wideband (UWB) radar, and multispectral remote sensing data, to reduce the influences of vegetation coverage on the estimation accuracy. A one-dimensional regression convolution neural network model is constructed to realize the quantitative prediction and estimation of SMC in farmland. The carried out experiments show that the proposed CNNR model has a better performance than traditional SVR and GRNN models and the R2, RMSE and RPD are 0.7453, 0.0140 cm3/cm3 and 1.9246, respectively. After introducing NDVI, MSAVI and DVI vegetation indices generated from multispectral images, the accuracy of the three models increased significantly. Among the three models, the constructed CNNR model has the best performance, and its R2, RMSE and RPD reach 0.9168, 0.0089 cm3/cm3, and 3.0201. Furthermore, after adding different levels of Gaussian noise to the original radar echoes, the CNNR model constructed in this paper still has the highest prediction accuracy and the strongest noise robustness. [ABSTRACT FROM AUTHOR]

• A detection approach for late-autumn shoots of litchi tree based on UAV remote sensing is provided. • A remote sensing dataset of late-autumn shoots of litchi tree is constructed. • A deep learning model based on the combination of YOLOv5 and Swin-transformer is proposed. • The detection accuracy of late-autumn shoots in remote sensing images is improved. • It will provide technical support for other researchers on late-autumn shoots and orchards manager. Litchi is one of the most common economic fruits in southern China, however, the growth of late-autumn shoots of litchi hinders flower bud differentiation and reduces yield of fruit. The early identification of the late-autumn shoots is of great significance for orchard management to control shoots and then increase fruit yield. At present, the identification of late-autumn shoots still relies on manual methods, which is not suitable for smart orchard management in a large area due to low recognition efficiency and high subjectivity. Therefore, a convenient, fast and cost-effective method is urgently needed. In response to this problem, the paper proposes a method based on the combination of unmanned aerial vehicle (UAV) remote sensing and object detection algorithm to detect late-autumn shoots. For this purpose, a remote sensing dataset of late-autumn shoots of litchi is first constructed by UAV. An improved YOLOv5 algorithm called YOLOv5-SBiC is then developed for late-autumn shoots identification. In the YOLOv5-SBiC algorithm, the transformer module is introduced to speed up the convergence of the network and improve detection accuracy, the attention mechanism module is employed to help the model extracting details, and BiFPN is used to better solve the multi-scale problem in detecting and then improve the recognition effect of small-sized objects. In addition, CIOU is selected as the loss function of bounding boxes regression to achieve high-precision localization of the boxes. The test results demonstrate that the recognition accuracy of YOLOv5-SBiC reaches a relatively high value of 79.6%, which is 4.0% higher than that (75.6%) of the original YOLOv5 algorithm and 15.9% higher than that (63.7%) of the pure transformer algorithm. It's also demonstrated that YOLOv5-SBiC is more competitive than the mainstream target detection algorithms in the dataset of late-autumn shoots. [ABSTRACT FROM AUTHOR]

• A more effective extraction method of time factor for soil type mapping based on remote sensing images was determined. • The method of principal component extraction result based on the tasseled cap transformation was more effective. • The fusion of multi-temporal features and texture features could improve the accuracies of soil type mapping. • The best texture feature for soil mapping was correlation parameter, which produced the largest overall accuracy of 86.81%. With the rapid development of remote sensing (RS) technology, remote sensing images provide an important data basis for soil type mapping. In remote sensing images, temporal factor is difficult to obtain directly, and the rich geometric features are not used adequately. Multi-temporal remote sensing data could effectively reflect the temporal variation of ground objects, while how to extract multi-temporal image features more effectively for soil type interpretation needs to be studied. Moreover, it is not clear whether multi-temporal features and texture features can be effectively integrated to improve mapping accuracies. Therefore, taking five soil types of Laoshan County, Shandong Peninsula, China as the subject investigated and six remote sensing images as data sources, this paper explored and compared two extraction methods of multi-temporal features from remote sensing images. The effects of the eight different texture features fused the multi-temporal features on digital soil mapping were also analyzed. The results showed that the principal component extraction result based on the tasseled cap transformation was better than based on the spectral band synthesis, increasing the overall accuracy by 3.83–11.41% and the kappa index by 0.03–0.13. The fusion of multi-temporal features and texture features can effectively improve accuracies of soil type mapping. After the addition of correlation texture feature parameter, the overall accuracy (86.81%) and Kappa index (0.81) were increased by 11.92% and 0.16, respectively. The research results showed that multi-temporal features in remote sensing images had great advantages in digital soil mapping, and the effective fusion with texture features provided a new idea for improving the accuracy of digital soil mapping. [ABSTRACT FROM AUTHOR]