Visual Interpretation Method Research Articles

Assessment of Landuse and Landcover Changes and its Implications to Flooding Along Omambala Flood Plain, Anambra State

This paper assessed the landuse and landcover changes and its implications to flooding along Omambala floodplain in Anambra State. The paper aimed at identifying the landuse and landcover types in Omambala floodplain in relations to their temporal dynamics, the extent of landuse change and its implication to recurring incidence of flood disaster in the study area over a period of 20 years starting from the year 1998 to 2018. Multi-temporal Landsat TM and ETM+ imageries were obtained at 10 years interval; 1998; 2008 and 2018 to observe the pattern of landuse and landcover along the floodplain over the period of study. Population data of study area were obtained from National Population Commission on four time paints as to understand the rate of land conversion and modification of the floodplain. Visual interpretation method was used to map landuse and landcover into five classes namely: waterbody; built-up; sparse vegetation; bare surface and dense vegetation. The result of the analysis of the imageries showed a considerable change in the pattern of landuse and landcover classes within the period under study. There was also a considerable population growth leading to progressive conversion of natural vegetation to other human activities with Built-up lands sparse vegetation and bare surface on the increase as well as their attendant consequences. However, the study concluded that various anthropogenic landuse activities especially poor farming system leading to increasing sparse vegetation, bare surface and built-up area are major factors that affects natural vegetation thereby worsening the incidence of flooding in the study area. The research therefore recommends relocation of settlements, greening of space already turning bare, effective floodplain management and continues monitoring of changes on land characters to ensure environmental sustainability and to forestall recurring incidence of flooding along the floodplain.

Evaluation of a visual interpretation method for tau-PET with 18F-flortaucipir.

IntroductionPositron emission tomography targeting tau (tau‐PET) is a promising diagnostic tool for the identification of Alzheimer's disease (AD). Currently available data rely on quantitative measures, and a visual interpretation method, critical for clinical translation, is needed.MethodsWe developed a visual interpretation method for 18F‐flortaucipir tau‐PET and tested it on 274 individuals (cognitively normal controls, patients with mild cognitive impairment [MCI], AD dementia, and non‐AD diagnoses). Two readers interpreted 18F‐flortaucipir PET using two complementary indices: a global visual score and a visual distribution pattern.ResultsGlobal visual scores were reliable, correlated with global cortical 18F‐flortaucipir standardized uptake value ratio (SUVR) and were associated with clinical diagnosis and amyloid status. The AD‐like 18F‐flortaucipir pattern had good sensitivity and specificity to identify amyloid‐positive patients with AD dementia or MCI.DiscussionThis 18F‐flortaucipir visual rating scheme is associated with SUVR quantification, clinical diagnosis, and amyloid status, and constitutes a promising approach to tau measurement in clinical settings.

CHIP: Channel-Wise Disentangled Interpretation of Deep Convolutional Neural Networks.

With the increasing popularity of deep convolutional neural networks (DCNNs), in addition to achieving high accuracy, it becomes increasingly important to explain how DCNNs make their decisions. In this article, we propose a CHannel-wise disentangled InterPretation (CHIP) model for visual interpretations of DCNN predictions. The proposed model distills the class-discriminative importance of channels in DCNN by utilizing sparse regularization. We first introduce network perturbation to learn the CHIP model. The proposed model is capable to not only distill the global perspective knowledge from networks but also present class-discriminative visual interpretations for the predictions of networks. It is noteworthy that the CHIP model is able to interpret different layers of networks without retraining. By combining the distilled interpretation knowledge at different layers, we further propose the Refined CHIP visual interpretation that is both high-resolution and class-discriminative. Based on qualitative and quantitative experiments on different data sets and networks, the proposed model provides promising visual interpretations for network predictions in an image classification task compared with the existing visual interpretation methods. The proposed model also outperforms the related approaches in the ILSVRC 2015 weakly supervised localization task.

Extrema of Elastic Properties of Cubic Crystals

As a rule, a discussion of physical properties of crystals is accompanied by far from simple mathematical calculations based on algebraic expressions in tensor and matrix notations. Such an approach caused by the nature and uniform presentation of properties of crystalline materials makes practical calculations of their specific characteristics and parameters very difficult. By an example of the titanium nickelide crystals, it is shown that the anisotropy coefficient of elastic properties of the cubic syngony crystals, which is equal to the ratio of the extreme values (minimum and maximum) of the shear modulus, is close to the ratio of the extreme values of Young's modulus. Some variants of describing the elastic anisotropy of cubic crystals using a series of dimensional and dimensionless independent indicators are considered. It is shown on a concrete example that they can give significantly different results. Methods of visual interpretation of the elastic properties anisotropy using the corresponding characteristic surfaces and their cross sections are discussed. It is noted that the characteristic surface of the Young's modulus of normal elasticity is the most accessible for construction, although it is not a complete characteristic of the anisotropy of elastic properties of cubic crystals. A method is proposed for visualizing matrices of elastic constants of crystals using the MatLab application package, which provides visual information on the ratio of the values of matrix elements. Single crystals of titanium nickelide TiNi, widely used in various fields of science, technology, and medicine and often discussed in the literature, are considered as an example of calculating the extreme values and parameters of anisotropy, as well as constructing characteristic surfaces and their cross sections.

Characteristics of Mainland Coastline Changes in Southeast Asia during the 21st Century

Song, Y.; Li, D., and Hou, X., 2020. Characteristics of mainland coastline changes in Southeast Asia during the 21st century. Journal of Coastal Research, 36(2), 261–275. Coconut Creek (Florida), ISSN 0749-0208.Based on multitemporal Landsat images covering the entire coastal zone, a visual interpretation method was adopted to extract the coastlines of mainland Southeast Asia (MSA) in 2000 and 2015 referencing Google Earth images and global distribution of mangrove data. Coastline structures, coastline fractal dimensions, coastline change rates, and patterns of land–sea interchange were analyzed to reveal the spatial–temporal characteristics of mainland coastline changes in the 21st century. The results showed that (1) during the past 15 years, the length and percentage of natural coastline decreased from 15,440.24 km (83.33%) to 14,909.75 km (78.91%) and artificial coastlines increased at an average growth rate of 29.04% from 3088.79 to 3985.89 km; (2) the fractal dimensions of most areas in MSA tended to increase under interference driven by human activities, and the coastline morphology became more complex overall; (3) coastlines had obvious evolution, in which 18.54% of the coastline advanced toward the sea and 11.25% retreated toward land at average rates of +20.36 and –15.41 m.a–1, respectively; and (4) the number of patches representing land–sea interchange reached 6607 so that the net land expansion area was 534.25 km2 and among these regions, those larger than 1 km2 accounted for only a small percentage (2.54%) of the total number but 79.41% of the total area. These hot spots were mainly concentrated in Myanmar, followed by Vietnam and Malaysia. Overall, the coastline of MSA was characterized by rapid artificial growth but still had a high percentage of naturalization. When it comes to developing the marine economy and addressing the ecological risk, the intensity of utilization and hardening of the coastline will likely increase.

Geosystems of the northeastern Hovsgol region

We examine the main natural features and the present status of geosystems across the territory of the northeastern Hovsgol region. The source of observational evidence concerning the state of the landscapes was provided by field expedition-based investigations, with emphasis placed on the study of the species and age composition of vegetation in conjunction with dynamical processes as well as geomorphological characteristics of the study area. The present status of geosystems across the study territory was analyzed by methods of visual interpretation and automated processing of spatial information using GIS technologies on the basis of Earth remote sensing data, field investigations of landscapes, topographic maps as well as literature sources. Special attention was paid to the analysis of the digital elevation model that was generated on the basis of SRTM 4 data in order to determine the morphometric characteristics of the location of geosystems, assign them to a particular type of classification of geosystems and compile the map legend. The legend to the landscape map was produced on the basis of a classification of a number of geomers as developed by V.B. Sochava in his theory of geosystems. As a result of the work done, we compiled the medium-scale landscape-typological map of geosystems (Sc 1:200000) on the level of groups of facies.

Bletilla striata planting area in Ningshan county extraction based on multi-temporal remote sensing images

Traditional Chinese medicine is planted in mountainous areas with suitable natural conditions. The planting area is complex in terrain,and the planting plots are mostly irregularly shaped. It is difficult to accurately calculate the planting area by traditional survey methods. The method of extracting Chinese herbal medicine planting area combined with remote sensing and GIS technology is of great significance for the rational development and utilization of traditional Chinese medicine resources. Taking Bletilla striata planting in Ningshan county of Shaanxi province as an example,the extraction method of planting area of traditional Chinese medicine in county was studied. High-resolution ZY-3 and GF-1 multi-spectral multi-temporal remote sensing images were used as data sources. Through field sampling,samples such as B. striata,cultivated land,forest land,water body,artificial surface,alpine meadow,etc. are collected. The spectral features,texture features and shape features of remotely identifiable objects in different planting areas and cultivated land,vegetable sheds were analyzed,confusing ground objects were eliminated and interpretation marks were establish. The method of visual interpretation is used to realize the extraction of B. striata planting areas,and the B. striata planting area are calculated by combining GIS technology. The results showed that the method of visual interpretation,using high-resolution ZY-3 and GF-1 multi-spectral multi-temporal remote sensing image data extracted the planting area of 403.05 mu. It can effectively extract the B. striata planting area in research region.

Remote sensing extraction method of Codonopsis pilosula planting area in Weiyuan county

Due to the large amount of Codonopsis pilosula planted in Weiyuan county,and the arable land area,the local medicinal materials office uses a large amount of manpower,financial resources and material resources to estimate its area every year. In order to extract the information of local Chinese medicinal materials more quickly and simply,we try to apply remote sensing technology to the extraction of Chinese medicinal materials. This paper will use Weiyuan county of Gansu province as the research area,and use the domestic ZY-3 Satellite multi-spectral remote sensing image as the data source to find out the spectral characteristics of the party's participation in other remote sensing images. The visual interpretation method was used to extract the planting area of the C. pilosula in Weiyuan county. The estimated value of the planting area of C. pilosula using satellite remote sensing technology was 75 965 mu( 1 mu≈667 m2),which was basically consistent with the field survey data of the local medicinal materials office. After the accuracy verification,it was found that the precision of C. pilosula planted by visual interpretation was more than 70%. It is concluded that the satellite remote sensing technology can be used to extract the information of C. pilosula and it can provide the relevant information of the planting area of Chinese medicinal materials quickly and accurately.

Extraction of distribution information of Angelicae sinensis plants in Weiyuan county based on remote sensing technology

Due to the large amount of nutrients required during the cultivation of Angelica sinensis and in order to prevent the occurrence of pests and diseases,and the annual reduction of the planting area of Angelica and the balance of supply and demand of A. sinensis,the A. sinensis plantation adopts the rotation mode. This paper takes Wuyuan county of Gansu province as the research scope and use GF-1 Satellite data as the data source,using remote sensing technology combined with field survey results,to explore the effective method of visual interpretation for the extraction of A. sinensis planting area. A sample was selected to generate a spectrum according to different feature types. The different characteristics of A. sinensis and other features were analyzed and distinguished in remote sensing images,so that the A. sinensis planting plots were extracted and verified in remote sensing images. The results showed that the accuracy verification value of the visual interpretation method was 95. 85%. It is determined that the visual interpretation method can effectively extract the A. sinensis planting plots within the research scope and realize the comprehensive grasp of the spatial distribution information of A. sinensis.

Floating Raft Aquaculture Area Automatic Extraction Based on Fully Convolutional Network

Cui, B.-G.; Zhong, Y.; Fei, D.; Zhang, Y.-H.; Liu, R.-J.; Chu, J.-L., and Zhao, J.H., 2019. Floating raft aquaculture area automatic extraction based on fully convolutional network. In: Jung, H.-S.; Lee, S.; Ryu, J.-H., and Cui, T. (eds.), Advances in Remote Sensing and Geoscience Information Systems of Coastal Environments. Journal of Coastal Research, Special Issue No. 90, pp. 86-94. Coconut Creek (Florida), ISSN 0749-0208.In the extraction of floating raft aquaculture areas from remote sensing images, the method of visual interpretation is time-consuming and laborious, and the traditional machine learning method has poor generalization ability and fitting ability for remote sensing data. To overcome these two problems, this paper proposes a method based on using a fully convolutional neural network to automatically extract floating raft aquaculture areas. The proposed method uses multiple convolution layers, pooling layers and nonlinear ReLU functions to extract nonlinear and invariant deep features of floating raft aquaculture areas, which effectively improves the accuracy of recognizing floating raft aquaculture areas. At the same time, L2 regularization and dropout strategies are added to the neural network model to avoid overfitting. The offshore area of Lianyungang in China was selected as the research region of the floating raft aquaculture. The experimental results show that the proposed method area in this paper effectively identifies and extracts the aquaculture areas.

Study of the Behavior of Super Resolution on Soft-Classified Output

Once the satellite sensor is in orbit, no hardware enhancement of the lens assembly can be done to improve spatial and spectral resolution. Super resolution (SR) as single frame or multi-frame can solve this problem up to a large extent. In this study, single- and multi-frame SR techniques were applied and tested on Worldview-2 datasets as well as on across-spatial datasets of LISS III and LISS IV. Study of soft classifier’s behavior on super-resolved images was performed through possibilistic C-means classifier. Quantitative methods based on calculation of peak signal-to-noise ratio, mean square error, root means square error, image quality index and qualitative methods of visual interpretation proved that both super-resolution methods remove outliers in an efficient way and resulted in images containing sharp edges. The single-frame super-resolution technique was found relatively inferior in terms of contrast and spatial resolution. Overall, multi-frame SR method outperformed other methods.

Analysis of Rice Field Area Conversion in Sleman Regency from 2000 to 2015, Using High-Resolution Satellite Imagery (Case Study: Ngaglik, Mlati and Depok Sub-District)

The purpose of this research is to find out how much area of rice fields which is reduced due to being converted into built-up land in Sleman Regency, especially in sub-districts which adjacent to Yogyakarta City, such as Depok Sub-district, Mlati Sub-district and Ngaglik Sub-district, from 2000 to 2015. Classification method used in this research is visual interpretation method which utilized on-screen digitization. The output of this research is a Map of Rice Field Conversion into Built-up Land at Depok, Mlati and Ngaglik Sub-district in Sleman Regency from 2000 to 2015. The results of this research prove that GIS can be used to determine the extent of changes in a rice field at Ngaglik, Depok and Mlati sub-districts. The area of rice field that was converted into built-up land in the research area is 864.45 ha.

Application of Full Polarimetric ALOS PALSAR for Land Cover Mapping In Sleman District

The simplest way to interpret polarimetric imagery for land cover classification is to use visualinterpretation methods. The existence of interpretations key as a tool for visual interpretation becomesimportant when different interpreters can produce different results. The quality of the results of theinterpretation of land cover is then determined by the quality of the interpretation tool, in this case, thekey to the interpretation of land cover. The purpose of this study was to make the key to land coverclass interpretation in the Full Polarimetric ALOS PALSAR image, then the interpretation key wasused for reference in making land cover maps and measuring the accuracy of the results of the visualinterpretation. The image used in this study consisted of HH, VV, HV and VH bands. The location ofthe study was in parts of Sleman District. The analysis is done visually by on-screen digitizing onALOS Palsar composite HH + VV HV + VH HH-HV image, which is then interpreted key. The truetest is done by means of the overall accuracy test and Kappa. Visually, ALOS PALSAR imagery isable to distinguish 12 land cover classes in the research area, namely built land, rice fields, mixedgardens, moorlands, salak garden, grass, forest, shrubs, open land, airports, water bodies and lavawith 83% Overall accuracy, and 78% Kappa accuracy.

Integration of Multiple Spectral Indices and a Neural Network for Burned Area Mapping Based on MODIS Data

Since wildfires have occurred frequently in recent years, accurate burned area mapping is required for wildfire severity assessment and burned land reconstruction. Satellite remote sensing is an effective technology that can provide valuable information for wildfire assessment. However, the common approaches based on using a single satellite image to promptly detect the burned areas have low accuracy and limited applicability. This paper develops a new burned area mapping method that surpasses the detection accuracy of previous methods, while still using a single Moderate Resolution Imaging Spectroradiometer (MODIS) sensor image. The key innovation is integrating optimal spectral indices and a neural network algorithm. We used the traditional empirical formula method, multi-threshold method and visual interpretation method to extract the sample sets of five typical types (burned area, vegetation, cloud, bare soil, and cloud shadow) from the MODIS data of several wildfires in the American states of Nevada, Washington and California in 2016. Afterward, the separability index M was adopted to assess the capacity of seven spectral bands and 13 spectral indices to distinguish the burned area from four unburned land cover types. Based on the separability analysis between the burned area and unburned areas, the spectral indices with an M value higher than 1.0 were employed to generate the training sample sets that were assessed to have an overall accuracy of 98.68% and Kappa coefficient of 97.46%. Finally, we utilized a back-propagation neural network (BPNN) to learn the spectral differences of different types from the training sample sets and obtain the output burned area map. The proposed method was applied to three wildfire cases in the American states of Idaho, Nevada and Oregon in 2017. A comparison of detection results between the new MODIS-based burned area map and the reference burned area map compiled from Landsat-8 Operational Land Imager (OLI) data indicates that the proposed method can effectively exploit the spectral characteristics of various land cover types. Also, this new method can achieve higher accuracy with the reduction of commission error (CE, >10%) and omission error (OE, >6%) compared to the traditional empirical formula method. The new burned area mapping method could help managers and the public perform more effective wildfire assessments and emergency management.

Research on the Water Area Extraction in Suzhou Based on Remote Sensing

In order to accurately obtain the information and calculate the area of water, the traditional visual interpretation method was used to extract the information of water in Suzhou with the high-resolution remote sensing image. By on-the-spot sampling measurement verification, the precision of visual interpretation was proved to meet the practical requirements. The research in experimental area indicate that the visual interpretation method is operable. The water area of Suzhou was obtained as 3205 square kilometers, and the area is 994 square kilometers except for the area of Taihu lake and the Yangtze river.

UNDERSTANDING OF FIRE DISTRIBUTION IN THE SOUTH SUMATRA PEAT AREA DURING THE LAST TWO DECADES

Peat fire is an annual catastrophic event in Indonesia, particularly in South Sumatra region during the dry season. Several previous studies have suggested that anthropogenic forces (land cover) contribute to peat fires. Analysis of the hotspots distribution on land cover types is needed to address peat fires in the futures. This study is aimed to evaluate the distribution of hotspots on the land cover type of the South Sumatra peat area during the last two decades (1997-2016). Remote sensing technique with visual image interpretation method using multitemporal satellite images is used to identify the variations of land cover in the study area. The results showed that the peat area of South Sumatra in 2013 dominated by ferns/ shrub (58%) and only about 10% covered by peat swamp forest (PSF). The largest hotspot distribution during the observation periods was spread in land cover types of Ferns/ shrub (34.808 hotspots) and Industrial Plantation (7.223 hotspots). Note that the densest hotspot located in the industrial plantation with an average of 37 hotspots/ 100 km2 / year. Meanwhile, only a small percentage of the hotspots located in PSF and occurred consistently over the last five years of observation periods. Based on results, the South Sumatra peatland is very vulnerable to burn during the dry season in the future. The government as policy maker must protect the existence of PSF and keep the ferns/ shrub from the fires and subsequent land cover changes (Industrial plantation or small holder area).

ДЕШИФРИРОВАНИЕ ЛАВИННЫХ ОТЛОЖЕНИЙ НА ЛЕДНИКЕ БЕЗЕНГИ ПО КОСМИЧЕСКИМ СНИМКАМ СО СПУТНИКОВ SENTINEL-2

Purpose. The aim of the paper is to evaluate the interpretation properties of Sentinel-2 satellite images from the point of view of revealing avalanche deposits on glaciers. Scientific novelty. Despite the sometimes quite high significance of avalanche deposits in the glaciers nourishment, the assessment of the Sentinel-2 satellites images applicability for solving the problem of detecting avalanche deposits on glaciers has not yet been carried out. Methods. The paper describes the main methodological approaches of revealing by such images the boundaries of the glacier, avalanche paths and avalanche deposits. The methods of visual interpretation of images synthesized in various color renditions were used. Results. During the work, an assessment was made of the suitability of using the Sentinel-2 images in studies of the Caucasus valley glaciers avalanche nourishment, as well as the possibilities and limitations of using these images for interpreting avalanche deposits within the glaciers were described. The practical significance. The results will allow to more effectively evaluate the avalanche nourishment of glaciers.

Glacier facies of Vestfonna (Svalbard) based on SAR images and GPR measurements

The warming climate of the Arctic affects the mass budget of glaciers, and changes in the distribution of glacier facies are indicative of their response to climate change. The glacial mass budget over large land ice masses can be estimated by remote sensing techniques, but selecting an efficient remote sensing method for recognizing and mapping glacier facies in the Arctic remains a challenge. In this study, we compared several methods of distinguishing the facies of the Vestfonna ice cap, Svalbard, based upon Synthetic Aperture Radar (SAR) images and terrestrial high frequency Ground Penetrating Radar (GPR) measurements. Glacier zones as determined using the backscattering coefficient (sigma0) of SAR images were compared against GPR data, and an alternative application of Internal Reflection Energy (IRE) calculated from terrestrial GPR data was also used for differentiating the extent of glacier facies. The IRE coefficient was found to offer a suitable method for distinguishing glacier zones and for validating SAR analysis. Furthermore, results of analysis of fully polarimetric Phased Array type L-band Synthetic Aperture Radar (ALOS PALSAR) and European Remote Sensing Synthetic Aperture Radar (ERS-2 SAR) images were compared with the IRE coefficient classification. Especially promising method is H-α segmentation, where the glacier zone boundaries corresponded very well with both GPR visual interpretation and IRE classification results. The IRE coefficient's simplicity of calculation makes it a good alternative to the subjective GPR visual interpretation method, where results strongly depend on the operator's level of experience. We therefore recommend for GPR profiles to be used for additional validation of SAR image analysis in studies of glacier facies on the High Arctic ice masses.

Monitoring land use and land cover changes due to extensive gold mining, urban expansion, and agriculture in the Pra River Basin of Ghana, 1986–2025

AbstractThis study used the double cumulative curve and visual image interpretation methods for the selection of spatiotemporal Landsat data to evaluate the land degradation by anthropogenic activities in the Pra River Basin (PRB) of Ghana. Unsupervised and supervised classification procedures were used to map the land use and land cover (LULC) distribution from 1986 to 2016. Assessment of LULC showed that the PRB has been subjected to six different rates of land degradation in the years 1986, 2000, 2004, 2008, 2013, and 2016. This is due to increase in settlement, cropland, and mining activities to about 130%, 198%, and 304%, respectively. The Markov chain and cellular automation integrated model was successful in predicting LULC distribution in 2016 and the outcomes were comparable to the actual LULC for 2016. The projected LULC for 2025 showed that land degradation is significant in the western and the eastern parts where cropland and forest are, respectively, converted to mining activity. The northern, southern, and middle parts of PRB are expected to experience high settlement expansion, sedimentation in the rivers, and cropland expansion, respectively. The results will aid natural resources management, planning, and sustainable development at PRB. In addition, the research method serves as guideline for other related studies in an attempt to investigate, quantify, and project LULC change in forest ecological areas.

Generalized space-time classifiers for monitoring sugarcane areas in Brazil

Spatially and temporally accurate information on crop areas is a prerequisite for monitoring the multiannual dynamics of crop production. Satellite images have proven their high potential for mapping crop areas at large scales, even at the crop-species level, when a classifier is calibrated on the same image with reference data corresponding to the same period. For operational monitoring purposes, however, it is critical to develop generalized classification methodologies applicable to large scales and different years. Generalized classifiers were presented in this study as follows: a) simple cross-year calibration and application (M1); b) multiyear calibrations (M2); and c) map updating through change detection with multiyear calibrations (M3). These three methods were developed in a classical frame of object-based classifications for a time series of Landsat images with the Random Forest machine learning algorithm. Therein, we tested these methods for sugarcane classification in Sao Paulo state, Brazil, as sugarcane is an economically important crop that has developed substantially in the past decades. Eight years of sugarcane reference maps were used to calibrate and validate the classifiers at four different sites. The cross-year application of M1 provided a low average accuracy Dice coefficient (DC) of 0.84, while it was, on average, 0.94 for the classical same-year calibration. When the classifier was trained on a multiyear dataset (M2), the accuracies achieved average values of 0.91 in independent years. The map updating method M3 showed promising results but was not able to reach the accuracy of visual interpretation methods for detecting annual sugarcane land use change. The multiyear classifier M2 was applied to four contrasting sites and provided reliable results for new sites and years for sugarcane classification. Calibration of the machine learning algorithm on a multiyear dataset of standardized and gap-filled satellite images and reference data proved to give an accurate and space-time generalized classifier, reducing the time, cost and resources for mapping sugarcane areas at large scales.