علیرضا صفدری نژاد

Photometry is a well-known method for 3D reconstruction of objects using images taken in different lighting conditions. In this method, by knowing the light sources' direction, the normal vectors of the surface are recovered in a dense grid through the intensities recorded in the captured images. Each normal vector is then converted to the height difference in two orthogonal directions, and the simultaneous estimation of the heights for the dense grid is done by solving a system of linear, overdetermined and inconsistent equations. The miss-alignment of the coordinate system represents normal vectors and the dense grid frame of 3D reconstruction causes a systematic error in the estimation of the gridded heights map. Photometric self-calibration methods for determining the light sources’ direction are one of the causes of miss-alignments in object and surface normal vectors coordinate systems. In this paper, a sequential and iterative process is proposed to estimate and perform an appropriate rotation to the surface normal vectors. In each iteration of this method, a portion of the necessary rotation is identified in order to parallelize of the two object coordinate systems and surface normal vectors through fitting a geometric transformation to the estimated residuals of the 3D reconstruction process. The results of using the proposed method in various experiments have demonstrated a noticeable improvement in the precision and accuracy of 3D reconstruction.

One way to ensure food security is to produce strategic agricultural products on a large scale using industrial methods. Managing large-scale farms consistently and cohesively is a challenging task that requires the utilization of modern technologies. Crop anomalies refer to uncommon and limited factors during agricultural production, leading to localized differentiation in the crop cultivation process. Factors contributing to crop anomalies in agriculture include imbalances in soil nutrients and fertilizers, grazing during crop growth, pests, variations in soil texture and slope in pastures, weed growth, and drought. Detecting and remediating factors limiting crop growth in vast agricultural lands is difficult and these issues are often noticed at harvest time. This article suggests a solution for continuously monitoring of large agricultural fields by analyzing the time series of Sentinel-2 satellite images. The effectiveness of this solution in detecting various anomalies of farms, in agrarian areas has been demonstrated by the results. The proposed solution offers features such as timely diagnosis, the ability to monitor the continuation of irregularities, and the measurement of compensatory measures' effectiveness. The method has successfully identified over five types of anomalies in the selected farms, achieving a detection accuracy of 95.60%.

Due to the wide applications of hyperspectral images, economical and innovative imaging systems are developed to acquire such images. In order to use hyperspectral images, it is necessary to establish an accurate relation between the ground space and the image space, which needs numerous Ground Control Points (GCPs). This fact highlights the need for developing geometric corrections methods for any camera design. BaySpec OCI-F (400-1000 nm) is one of the innovative cameras that acquires pushbroom hyperspectral images. In addition to the pushbroom sensor, the camera uses a frame sensor that acquires images at the same time as the pushbroom sensor and with the same temporal rate. In this article, a geometric correction method for pushbroom images of OCI-F camera is proposed. Based on the camera’s imaging design, the first step of the method determines a set of calibration parameters which geometrically relates the pushbroom and the frame sensors. Then using this relation and the geometric relations among consecutive frames, the pixels of the pushbroom scene are rearranged and form the corrected image. The proposed method determines the relation among the consecutive images via Least Square Matching (LSM) method. The results show that the correction method has decreased the geometric distortions of the raw pushbroom scene by 62.2% on average. Such a reduction causes the average accuracies of two-dimensional and three-dimensional generic models which relate image space and ground space together, to increase by 34.1% and 39.9% respectively.

Analyzing the image blocks captured before and after geometrical changes is known as the conventional approach for detecting them in photogrammetric applications. Developed methods can be categorized into 1- comparison of 3D models generated via the image blocks and 2- direct comparison of single images. The occurrence of radiometric differences in the geometrically changed areas can increase their discrimination and facilitate their detection. However, the occurrence of geometric changes without sensible radiometric effects is a special type of change that its identification is faced with more challenges. Slight displacement of the objects in the scene, small landslides, subsidence or uplift, the effects of local pressure and tension on objects in the industrial procedures and etc. are some examples of geometric changes that do not have a noticeable radiometric appearance in the images.In the absence of incorrect observations, simultaneous triangulation of image blocks captured before and after geometric changes is a simple and effective way of reaching to detection of changes. In other words, by identifying the corresponding points in the fixed regions of the scene in the image blocks, the simultaneous triangulation of the image blocks captured in both epochs can align them in a unique object coordinate system. Thus, it can be possible to generate two independent and co-registered 3D models for identifying the occurred changes. However, maintaining the radiometric similarity of the changed areas leads to the identification of wrong-matched points when using automatic image matching methods.The inclusion of an unknown 3D position for each wrong-matched point in the changed areas leads to a defect in the design of the mathematical model for the bundle adjustment. These defects result in incorrect generation of the 3D models, large and systematic errors in the residuals of observations, and incorrect estimation of the extrinsic parameters of images. The remedy to this defect is to assign two distinct unknown 3D positions for each wrong-matched point before and after changes in the bundle adjustment. Lack of prior knowledge of the wrong-matched points located in the changed areas is the cause of this problem. In this article, an iterative solution is proposed to identify and correct the effects of the wrong-matched points in the process of simultaneous bundle adjustment.

In the proposed method, at first, all the confident radiometrically matched points among all images taken before and after the geometric changes are detected via the well-known feature-based image matching methods. Their matched positions, then, are again accurately rectified and verified by the least squares image matching method. The matched points identified after refinement are classified into two categories. 1- The matched points that have been detected only in the images of one image block and 2- The matched points that have been detected at least in two images in each image block. Among the points of the second category, there probably are matched points that are geometrically changed between two epochs, but their radiometric similarities have made to incorrectly identified as the matched points between two image blocks. In this paper, these were called the wrong-matched points which are iteratively identified and their corresponding mathematical models are corrected in the triangulation process.To do so, three different bundle adjustments are performed as the first step. Independent triangulation of the image blocks captured before and after the geometric changes and the simultaneous bundle adjustment of both blocks via the initially detected matched points of the first and second categories are the first three triangulations. Due to the existence of wrong-matched points, the initial simultaneous triangulation has a defect in the design of the mathematical model, which is gradually and in an iterative process, the wrong-matched points located in the changed areas would be identified.Identification of the wrong-matched points is done using the relative comparisons on their residual vectors. The comparisons are designed in two consecutive statistical tests. The main idea of this method has been inspired by the well-known Baarda test in the detection of gross errors in the observations of geodetic networks. By gradual identification of the wrong-matched points, their corresponding mathematical model will be modified in the bundle adjustment.To do so, the unknown values of the 3D coordinates of these points are separated for the time before and after the change epochs.This action by modification of the mathematical model in the bundle adjustments brings back the relative equilibrium in the estimation of the residual vector of observations.

Implementation and comparison of the proposed method with a conventional geometric approach in identifying the incorrectly matched points (using robust estimation of the epipolar geometry) have shown the adequacy and superiority of the proposed method. The proposed method, on average in more than 11 different experiments, was able to achieve an average accuracy of 85.8% in identifying the changed points. Meanwhile, the proposed method shows a 34.5% improvement compared to the conventional geometric approach based on epipolar geometry.       

Conclusions and suggestions:

The proposed method is an effective solution for identifying the geometrically changed points in the simultaneous triangulation of image blocks before and after geometric changes when the changed areas have a stable radiometric similarity. This method is more sensitive to the occurred changes than the conventional method of identifying incorrect correspondences based on epipolar geometry. Iterative adjustment of the observations’weight matrix through the Variance Components Estimation (VCE) techniques in order to detect and eliminate the effects of wrong-matched points can be considered a future research topic in this field.

Map of croplands is one of the information layers required in the efficient management of these lands. Having such maps makes it possible to monitor agricultural fields during the growing season continuously. In this study, a solution to produce map of Shahrekord’s agricultural lands in two agricultural and non-agricultural classes is presented using the time series of different extracted indices from Sentinel-2 images. Since the use of large data sources is one of the obstacles to the development of methods based on the time series of satellite images, the Google Earth engine processing platform has been used in this study. The proposed method is based on integrating supervised pixel-based classification results with segmentation results. First, training data of supervised classification is provided in a rigorous refining process without the need of collected data from field surveys or interpretation of high-resolution satellite images. Then, by calculating the separability of the two target classes in the time series of each index, the optimal indices are selected. Finally, by combining the results of segmentation and classification methods based on the votes obtained from the classification results, agricultural or non-agricultural class is assigned to each of the image segments. In addition to incorporating spatial information including edges and spatial proximity, this method has been able to improve the noise and porous results of pixel-based classification and has increased the overall accuracy of the final map from 90.7% to 96.05%. Also, user accuracy of both agricultural and non-agricultural classes show an improvement of 3.27 and 7.97%, respectively.

In this paper, a novel method has been proposed to automatically produce 3D videos through amateur digital cameras that have been fixed with each other. Asynchrony of the videos (frames per second rates and start times), unavailability of the exact camera internal parameters, and the technical limitation of precise relative adjusting the stereo cameras could be known as the main challenges of producing 3D videos via the amateur cameras. In the proposed method, the videos acquired by the stereo camera are synchronized through the automatic matching between temporal indices. Then, the calm periods of videos (the times with zero relative velocity between camera and scenes) are detected to be used for selecting proper matched image frames. Proper matched frames are then applied for finding matched points via a geometrical constrained feature-based matching method. The matched points are used for self-calibration as well as relative parameters estimation of the stereo cameras. In the last step, epipolar resampling procedure has been used to generate normalized videos. Automatic and precise synchronization of the stereo videos as well as the proper generalization of the proposed approach in the different sample datasets has been seen in the evaluation processes. Spectator satisfaction in the depth perception of the 3D videos is another quality achievement of the proposed method.

Image fusion is known as a synergetic process for merging multispectral and panchromatic images contents. So far, various methods have been developed in which the usage of the frequency domain is one of them. The frequency-based image fusion techniques are performed using high and low pass filters. So, the determination of the sizes of these filters would be a challenge. In this paper, a weighted index is proposed to determine the sizes and shapes of the low and high filters in fusion of the panchromatic and multispectral images. In the proposed method, the weights of the spectral and spatial indicators are independently estimated for each image. So, the effects of the differentiation of the image contents and different range of the indicators are properly adjusted to reach the optimum filtering. The comparison of the best results obtained from the proposed method with the other well-known fusion methods, in the used datasets, was indicated an average improvement of 58% in RMSEs.

Over the past three decades, with the rapid development of spatial-based satellite imagery, remote sensing technology has found a special place in various applications of urban management. Production of status maps of urban structures, the study of energy loss status, identification of thermal islands, monitoring of urban vegetation, and assessment of air pollution are just a few examples of areas related to urban management that remote sensing technology is the basis for indirect measurement of the related quantities. Maps of urban structures such as building blocks are commonly used in crisis management, urban design, and urban development studies.
 Materials

In this study, the production of urban building block maps using Sentinel 1 and 2 satellite images has been conducted. Normalized Difference Vegetation Index (NDVI) and Normalized Difference Building Index ( NDBI ) for three consecutive months, the slope feature derived from the 30-meter Shuttle Radar Topographic Mission (SRTM)Digital Elevation Model of the study area, along with two Vertical – Vertical (VV) and Vertical - Horizontal ( VH ) polarization in both ascending and descending orbits, form the set of input features.

The proposed method of this paper relies on the use of a generalizable trained classifier. Initially, the classifier is trained in 2015 using training samples obtained from a new rigorous refining process using different remote sensing and spatial products. This rigorous refining process uses a reference urban map of 2015. In the first step, the corresponding areas related to the ways and roads are removed using the OpenStreetMap data layer. Areas suspected of vegetation with NDVI greater than 0.2 are then discarded. Also, due to the high backscattering of buildings in Synthetic Aperture Radar images, areas with a value less than the average backscattering coefficient of the remaining areas are eliminated. Finally, the residual map is refined using the Mahalanabis distance and the Otsu automatic thresholding method. The trained classifier is then used to generate a map of building blocks at similar time intervals for the three target years (2018, 2019, and 2020). Due to the diversity of texture and density of building blocks in the metropolis of Tehran, the proposed method has been evaluated in this area. Due to the concentration of political, welfare, and social facilities, Tehran has experienced more unplanned and irregular expansion and urbanization than other cities in Iran, which has lead to changes in buildings and constructions. Also, due to the availability of free satellite images and various online processing operations, the Google Earth Engine platform has been used in this study. The performance of three different classifiers including Random Forest (RF), Minimum Mahalanabis Distance (MD), and Support Vector Machines (SVM) are examined in this process. In order to evaluate the proposed method, reference samples obtained from visual interpretation of high-resolution satellite images (Google Earth) in all three target years have been used.

The performance of the aforementioned classifiers has been investigated using 3 different criteria: overall accuracy, user accuracy, and F-score of building blocks. The RF method with an overall accuracy of over 93% in all three target years has shown the best performance. The SVM method ranks second with an accuracy of about 91% every three years. However, the MD method with an overall accuracy below 85% in all three target years has not performed well.

The results show better performance of the RF method in all three target years with an overall accuracy of over 93%. It should be noted that the MD classifier with higher user accuracy than other methods, has shown better performance in detecting the class of building blocks. However, the RF method is the best classifier in terms of the user accuracy of the background class. The effect of using two VV and VH polarization and also the slope derived from the SRTM Model in the input feature set on the final accuracy of classification was also investigated. According to the results, the simultaneous use of these three features produces more accurate results in both target classes. However, the results show that the use of VV polarization increases the final classification accuracy compared to VH polarization. The presence of slope feature along with both polarizations has also increased the classification accuracy of each class, especially the background class. However, the exclusion of both VV and VH features from the input feature set has resulted in a more than 10% reduction in overall classification accuracy.

Based on calculated overall accuracies which are above 80% in the majority of investigated cases, two different results can be concluded. First, the trained classifier has shown good temporal generalization and has achieved acceptable accuracy in the target years. Second, due to the different collection processes of training and evaluation data, the proposed rigorous refining method for the preparation of training data has shown good performance. The effect of using two VV and VH polarization and also the slope derived from the SRTM  Digital Elevation Model in the input feature set on the final accuracy of classification was also investigated. According to the results, the simultaneous use of these three features produces more accurate results in both target classes. However, the results show that the use of VV polarization increases the final classification accuracy compared to VH polarization. The presence of slope feature along with both polarizations has also increased the classification accuracy of each class, especially the background class. However, the exclusion of both VV and VH features from the input feature set has resulted in a tangible decreasein overall classification accuracy.

In this paper, a simple structured light system is designed to produce the three-dimensional points cloud from the un-textured surfaces. The system consists of two cameras and a planer laser, in which the 3D contents are produced through the stereo images taken from the light reflected by the intersection of a planer laser and the 3D surface of an object. There was no control over how the laser swept through the surface and the instantaneous parameters of the laser plane were not known in advance. Considering the knowledge of the internal camera calibration parameters and the relative orientation of the stereo-pairs, the video captured by the cameras are normalized during the epipolar re-sampling process. Next, in each pair of simultaneous frames, the matched points located at the 3D section of the laser’s plane are then identified. During the simultaneous space intersection of the matched points, a constraint is applied to enforce the singularity of the covariance matrix of 3D points lie in the intersection of the laser's plane and the 3D surface of an object to ensure their co-planarity. By applying this statistical constraint, the precision of the surface 3D reconstruction was improved up to 41% in this structured light system.

The ability of recording the high resolution spectral signature of earth surface would be the most important feature of hyperspectral sensors. On the other hand, classification of hyperspectral imagery is known as one of the methods to extracting information from these remote sensing data sources. Despite the high potential of hyperspectral images in the information content point of view, there are a numerous challenges in reliable extraction of information from these images. The issues such as 1- spectral similarity of different phenomena, 2- sensor noises and atmospheric effects, 3- the effects of high dimensionality in the pattern recognition algorithms, 4- the necessity of large number of training data to perform a reliable classification, and 5- spectral variability of similar phenomena could be considered as some of the challenges in hyperspectral data processing. Decreasing of the high dimensionality effects via the dimension reduction algorithms (e.g. band selection and feature extraction algorithms), as well as increasing the separability of the overlapped classes through the linear/non-linear mappings into the feature spaces with the higher dimensional are two opposite and conventional approaches of hyperspectral data processing. These approaches would be used based on the factors such as 1- complexities of classes in the imaging area, 2- spectral range of imaging sensor, and 3- the restrictions of processing algorithms. In this paper the fusion of these two approaches is used to perform an accurate hyperspectral image classification. To do so, a novel feature extraction method is proposed to be used in the hyperspectral image classification. The core of this method is the fusion of the linear, non-linear and sparse representation based features which is used to produce the effective features in the weighted K-Nearest Neighbors (KNN) classification method. In this procedure, a set of supervised and nonlinear features are extracted as the first step through the Nonlinear Principal Component Analysis (NLPCA). The supervised usage of NLPCA in order to extract features is known as one of the novelties of this paper. In this step, the spectral bands are usually mapped to a high dimensional feature space through the self-estimator artificial neural networks (ANNs) which are trained separately by ground truth data. In the second step, the previously extracted features are linearly transformed by the Linear Discriminate Analysis (LDA) method in order to reduce the dimension of the hypercube generated via supervised NLPCA to a separable feature space. In the last step, a set of features which is proportional to the number of classes is generated based on the sparse representation theory. The sparse representation features were hired to handle the effects of the inter-class variability. The precisions of the classified features in the two different hyperspectral images were on average shown 6 percent improvements in comparison with the spectral bands and the other combinations of extracted features. Furthermore, reach to the approximately 99% overall accuracies in the classes with the few training data could be considered as other achievements of the proposed method.

One of the consequences of a fire is smoke. Occasionally, monitoring and detection of this smoke can be a solution to prevent occurrence or spreading a fire. On the other hand, due to the destructive effects of the smoke spreading on human health, measures can be taken to improve the level of health services by zoning and monitoring its expansion process. In this paper, an automated method is proposed to detect the dilute smoke caused by large fires in multispectral images. The main idea of this method is the impossibility of precisely reconstructing the smoke in the bands affected by smoke (blue band) using regression models from other spectral bands. In the first step of the proposed method, the absolute value of the residuals of the regression estimation of blue spectral band is transformed into a binary mask with the help of Otsu thresholding. Afterwards, in an iterative process, non-smoke areas are detected and then clustered. In the iteration process, a regression model is fitted for each cluster and for each pixel, coefficients with the least error of the blue band reconstruction is used. Through more accurate estimation of the blue band, it reduces the effect of First Positive Error and leads the mask of residuals obtained from thresholding process to the smoke areas. The final step of the proposed method is to refine and remove the incorrect image segments. This method has been successful in detecting diluted smokes and also in disregarding smoke in non-smoky images. The results show the average accuracy of  99.04 percent in several datasets with diluted smokes.

High accuracy and huge density of 3D points cloud acquired by airborne Lidar makes them as a good and suitable tool in order to analyze of terrain surface. In this procedure, points cloud clustering is a fundamental step in the procedure of information extraction form LiDAR's data. In this paper a novel method is proposed for supervised classification of LiDAR points cloud based on contextual analysis on LiDAR points. The proposed method consists of three main steps. In the first step, a set of contextual features are produced for each points in LiDAR data. In second step, optimum feature selection is done in the modified prototype space using a new strategy. The last step is conducted to a simple k-means clustering on the feature space spanned by optimum contextual clusters. An urban area with the residential texture has been used as the case study to evaluation of the proposed method. The results indicate proper classification accuracies. The overall accuracies and kappa coefficients was 93.15% and 0.89 respectively.

In normal images, which resampled according to epipolar geometry, all of spatial displacements of points in the space of stereo images occur only in one direction of the digital image coordinate system. This prominent characteristic makes normalized imagery as an important prerequisite for many photogrammetric activities such as image matching, automatic aerial triangulation, automatic digital elevation model and orthophoto generation, and stereo viewing. In this paper, a novel approach for epipolar resampling of linear pushbroom satellite imagery is proposed based on Orbital Parameters Model (OPM). The proposed method is developed based on modifying the exterior orientation parameters of OPM in the object space. The most prominent advantage of this method is the capability of the correction of off-nadir viewing of the sensor through the physical interpretation of its parameters. Also, there is the capability of implementation of the proposed method by means of other common physical or interpolative mathematical models used in geometric correction of satellite imagery. According to the results, the average reminded vertical parallax x in the digital image coordinate system is determined 0.73 pixels with respect to the independent check points that demonstrates the high performance of the proposed method.

Investigation of various types of vegetation’s characteristics as an effective parameter in the energy exchange between the atmosphere and Earth's surface is very important in environmental, natural resources and agriculture studies. Nowadays, using remote sensing techniques with a wide range of valuable spectral information facilitate the study of vegetation, especially in estimation of the biophysical parameters. One of the most important biophysical parameters used in the various analyses related to the study of vegetation is Leaf Area Index (LAI). In this study, in addition to the analyzing and modeling of the relationship between LAI and vegetation indices (VIs)via spectrometry observations, the limitations of the mathematical model for estimation of LAI has been explored, some practical guidelines have been provided to improve the accuracy of the model as well as a new vegetation index has been designed. Finally, the results showed that through the conventional vegetation index, Simple Ratio (SR) and Second Soil Adjusted Vegetation Index (SAVI-2) have the minimum RMSE (about 0.08 in LAI unit) and the fitted models using their formulas in comparison with the other indices have the minimum rate of saturation. In other words, these indices are more efficient to estimation of the LAI; especially in high density vegetation area and can be used with high reliability in linear models for LAI estimation.

This paper proposes a new approach for geometrical modeling of satellite imagery which uses 2D-polynomials for 3D point determination from satellite stereo images. In this model, 2D polynomials are considered as additional parameters in colinearity equation, instead of considering as models for relating between the ground and space images. Orbital parameters model are used as fundamental colinearity equations in this modeling. Essential parameters in the orbital parameters model are determined from satellite ephemeris data and they are considered as fixed parameters in the modeling. In this model, polynomial coefficients are the only unknown parameters which are determined from GCPs in a linear equations set. The major advantages of this model are: Decreasing the performance complexities in using orbital parameters model, ease of implementation, applicability on raw geometrically corrected images, and possibility of using maximum capacity of satellite auxiliary data, and linearity of equations in space intersection procedure. Implementation of this model on different datasets shows high potentiality of the mentioned approach for 3D point determinations from satellite stereo images.

Taking the advantages of polarimetric radar data has a decisive role in target detection purposes. In this way, comprehensive geometric and descriptive information could be derived through processing this kind of data. However, the selection of optimal features could be considered as a major challenge in order to classification of the polarimetric radar imagery. In this paper, a novel approach is proposed for optimal feature selection based on mapping the extracted features to the prototype space. As a key result of the paper, fitness index is introduced to facilitate the optimal feature selection in polarimetric radar images. On the other hand, the mixture of backscattering mechanisms in a pixel level is another limitation to obtain precise spatial information. Thus, utilizing soft classifiers is indispensible to acquire the sub-pixel information. Positivity and sum to unity of the fractions within each pixel are major challenges in results of the soft classifiers. In this paper, integration of the soft classifiers and unsupervised algorithms of end-member extraction is proposed to solve this problem. Likewise, soft classifiers just provide fractional maps and the spatial arrangement of sub-pixels remains unknown. In this regard, Super Resolution Mapping (SRM) techniques are developed to enhance the spatial resolution of the results of soft classifiers. This research attempts to provide a sub-pixel classification of polarimetric radar images using the pixel swapping technique. Towards this end, a non-random procedure is suggested for initial arrangement of the sub-pixels. According to the results, the proposed method for optimal feature selection is demonstrated more accurate results than genetic algorithm. Next, three algorithms including Linear Spectral Unmixing (LSU), Multi-Layer Perceptron (MLP) and Support Vector Machines (SVM) are performed to soft classifying of the polarimetric radar image into three classes (residential, vegetation and bare earth). SVM present accurate results in comparison to others; its resulted fractional maps are used in SRM procedure. Finally, pixel swapping technique is performed based on the results of SVM classification and the land cover map of the study area is produced in a finer spatial resolution.