جستجوی مقالات مرتبط با کلیدواژه "sentinel-2 satellite" در نشریات گروه "مکانیزاسیون کشاورزی"

Remote sensing is defined as data acquisition about an object or a phenomenon related to a geographic location without physical. The use of remote sensing data is expanding rapidly. Researchers have always been interested in accurately classifying land coverage phenomena using multispectral images. One of the factors that reduces the accuracy of the classification map is the existence of uneven surfaces and high-altitude areas. The presence of high-altitude points makes it difficult for the sensors to obtain accurate reflection information from the surface of the phenomena. Radar imagery used with the digital elevation model (DEM) is effective for identifying and determining altitude phenomena. Image fusion is a technique that uses two sensors with completely different specifications and takes advantage of both of the sensors' capabilities. In this study, the feasibility of employing the fusion technique to improve the overall accuracy of classifying land coverage phenomena using time series NDVI images of Sentinel 2 satellite imagery and PALSAR radar imagery of ALOS satellite was investigated. Additionally, the results of predicted and measured areas of fields under cultivation of wheat, barley, and canola were studied.

Thirteen Sentinel-2 multispectral satellite images with 10-meter spatial resolution from the Bajgah region in Fars province, Iran from Nov 2018 to June 2019 were downloaded at the Level-1C processing level to classify the cultivated lands and other phenomena. Ground truth data were collected through several field visits using handheld GPS to pinpoint different phenomena in the region of study. The seven classes of distinguished land coverage and phenomena include (1) Wheat, (2) Barley, (3) Canola, (4) Tree, (5) Residential regions, (6) Soil, and (7) others. After the preprocessing operations such as radiometric and atmospheric corrections using predefined built-in algorithms recommended by other researchers in ENVI 5.3, and cropping the region of interest (ROI) from the original image, the Normalized Difference Vegetation Index (NDVI) was calculated for each image. The DEM was obtained from the PALSAR sensor radar image with the 12.5-meter spatial resolution of the ALOS satellite. After preprocessing and cropping the ROI, a binary mask of radar images was created using threshold values of altitudes between 1764 and 1799 meters above the sea level in ENVI 5.3. The NDVI time series was then composed of all 13 images and integrated with radar images using the pixel-level integration method. The purpose of this process was to remove the high-altitude points in the study area that would reduce the accuracy of the classification map. The image fusion process was also performed using ENVI 5.3. The support Vector Machine (SVM) classification method was employed to train the classifier for both fused and unfused images as suggested by other researchers.To evaluate the effectiveness of image fusion, Commission and Omission errors, and the Overall accuracy were calculated using a Confusion matrix. To study the accuracy of the estimated area under cultivation of main crops in the region versus the actual measured values of the area, regression equation and percentage of difference were calculated.

Visual inspection of classified output maps shows the difference between the fused and unfused images in classifying similar classes such as buildings and structures versus regions covered with bare soil and lands under cultivation versus natural vegetation in high altitude points. Statistical metrics verified these visual evaluations.  The SVM algorithm in fusion mode resulted in 98.06% accuracy and 0.97 kappa coefficient, 7.5% higher accuracy than the unfused images.As stated earlier, the similarities between the soil class (stones and rocks in the mountains) and manmade buildings and infrastructures increase omission error and misclassification in unfused image classification. The same misclassification occurred for the visually similar croplands and shallow vegetation at high altitude points. These results were consistence with previous literature that reported the same misclassification in analogous classes. The predicted area under cultivation of wheat and barley were overestimated by 3 and 1.5 percent, respectively. However, for canola, the area was underestimated by 3.5 percent.

The main focus of this study was employing the image fusion technique and improving the classification accuracy of satellite imagery. Integration of PALSAR sensor data from ALOS radar satellite with multi-spectral imagery of Sentinel 2 satellite enhanced the classification accuracy of output maps by eliminating the high-altitude points and biases due to rocks and natural vegetation at hills and mountains. Statistical metrics such as the overall accuracy, Kappa coefficient, and commission and omission errors confirmed the visual findings of the fused vs. unfused classification maps.

Introduction :

Achieve more production, efforts should be made to increase yield per hectare. One of the things that play an important role in increasing crop production, disease control, chastity control, improving soil fertility and structure is the implementation of proper crop rotation. Crop rotation increases the efficiency of production and yield through the continuity of soil vegetation, more efficient water use, preservation of soil nutrients, increase of soil organic matter and stability of soil grains, reduction of pests and diseases, and better control of weeds. Also, data collection in the conducted research is done in a traditional way, which is usually difficult, limited and very time-consuming due to the dispersion of farms and their size.

Materials and Methods:

 His study was conducted to investigate the effect of crop rotations on wheat yield using satellite images in three crop years 2017-2018, 2019-2028, 2019-2020 in the fields of Shavor region of Khuzestan province. In this research, all the evaluated images are related to the Sentinel-2 satellite and all these images were obtained from the US Geological Survey website. The satellite images were taken at the flowering stage of wheat, and images without clouds and fog were used on February 25, 2019 (for the year before rotation) and February 19, 2021 (for the year after rotation). Also, for pre-processing and processing and extracting information, SNAP software, Sen2Cor and ENVI plugin were used, respectively. The steps of this research were done in three steps. In the first stage, five plant spectral indices EVI, GNDVI, GARI, NDVI and RVI were evaluated to identify the best index to estimate wheat yield. The spectral index, which has a higher correlation with the yield of wheat, was chosen as the base index and was used to continue the research. In the second stage, three farms were randomly selected from each rotation to evaluate wheat yield after their application. In this section, variance analysis was performed in the form of a completely random design in three replications (one replication for each farm). The treatments include four alternations of wheat-wheat-wheat, wheat-canola-wheat, wheat-rice-wheat and wheat-clover-wheat. At this stage, the comparison of means was done by Duncan's multi-range test and in the MSTATC software environment. The third stage is the changes in wheat yield in each rotation in two times before and after applying that rotation. For this purpose, the changes of the base spectral index before and after the application of periodicity were set as criteria.

Results:

The results of variance analysis of five spectral indices studied in this research showed that the coefficient of explanation of each of these indices with wheat yield at the time of flowering is NDVI with 76, RVI with 73, GARI with 71, EVI with 60 and GNDVI with 57 respectively. In this research, the NDVI spectral index has the highest correlation, R2 of 76%, and the minimum error, RMSE of 0.547 earned the results showed that the average and the dispersion coefficient of the NDVI spectral index of intervals have a significant difference at the probability level of 1%. So that in terms of the average, the lowest average of the NDVI spectral index is in wheat-rice-wheat rotation with a rate of 0.2650 and the highest average is in the wheat-clover-wheat rotation with a rate of 0.5603. According to the distribution coefficient, the minimum and maximum values belonged to the rotation of wheat-canola-wheat with the rate of 0.0505 and wheat-rice-wheat with the rate of 0.1970. The results of the corresponding comparison before and after the application of each rotation showed that not observing the rotation and wheat cultivation after two crop years led to a 10% decrease and the use of rice in the crop rotation led to a 50% decrease in the NDVI spectral index. Also, the use of rapeseed and clover in crop rotations has led to an increase of 2 and 30% in NDVI spectral index compared to before rotation. The results of the dispersion coefficient of the NDVI spectral index in the time before and after the application of rotation showed that in the continuous cultivation of wheat, the dispersion coefficient due to the decrease in yield uniformity in different parts of the field led to an increase of 27% and in the rotation of wheat-rice-wheat it led to an increase of 152 became a percentage However, the distribution coefficient of wheat-canola-wheat rotation and wheat-clover-wheat rotation resulted in a decrease of 57 and 32%, respectively, due to the increase in yield uniformity in different parts of the field.

Conclusion:

 Heat is one of the strategic products, and the evaluation of different rotations is of particular importance in increasing its yield. In this research, five plant spectral indices EVI, GNDVI, GARI, NDVI and RVI were investigated in order to identify the base index for wheat yield estimation. The results of the analysis of these indices showed that the NDVI spectral index with an explanation coefficient of 76% has the highest correlation with wheat yield. The comparison results of the NDVI spectral index correspondingly in each rotation in two states before and after the rotation showed that the continuous cultivation of wheat in an agricultural land after two crop years led to a 10% decrease in the NDVI spectral index and the use of rice in the wheat-rice rotation. - Wheat leads to a 50% decrease in the NDVI spectral index of wheat; But the use of canola and clover in the rotation of wheat-canola-wheat and wheat-clover-wheat led to an increase of 2% and 30% of NDVI spectral index, respectively. Also, the results of the comparison of the dispersion coefficient of the NDVI spectral index before and after the application of rotation showed that in the continuous cultivation of wheat, the dispersion coefficient increased by 27% due to the decrease in yield uniformity in different parts of the field, and in the wheat-rice-wheat rotation, the dispersion coefficient also as a result of the reduction of yield uniformity in different parts of the farm, it led to an increase of 152%.