Mohammad Kamal
- Courses7
- Reviews20
- School: University of Saskatchewan
- Campus:
- Department: Geography
- Email address: Join to see
- Phone: Join to see
-
Location:
Saskatoon, SK - Dates at University of Saskatchewan: April 2010 - July 2016
- Office Hours: Join to see
Biography
University of Saskatchewan - Geography
GIS/Remote Sensing Specialist
Mohammad Mostafa
Kamal
Saskatoon, Saskatchewan, Canada
As a Geographer with advanced educations and trainings in the Geographic Information Sciences (GIScience), and technologies, my primary expertise and work interest is in the area of integrated applications of Geographic Information System (GIS), Remote Sensing (RS) and Global Position System (GPS) tools and techniques in geophysical environmental observation, assessment, changed detection, monitoring and modeling for prediction.
I have 15+ years of experience in Geo-spatial Technology through working for Government sector projects, international donor and development agencies, consulting companies, NGOs, and teaching in post-secondary institutions in Bangladesh, Thailand, The Netherlands, England, and Canada.
Experience
University of Saskatchewan
Assistant Professor of Geomatics and Hydrology
Teaching
Undergraduate and Graduate research supervision
Other Academic services
Education
Asian Institute of Technology, Thailand
M.Sc.
Remote Sensing and GIS
Thesis: A GIS and Remote Sensing Based Approach for Evaluation of Geophysical Environmental Consequence of the Road System in Phetchaburi province in Thailand This study used GIS and remote sensing and statistical methodologies for finding and quantifying the relationship between the expansion of road network through the forest area and the rate of deforestation, and the changes in localized flood regime with road concentration. The study observed a significant association between them.Middlesex University, London, UK.
Doctor of Philosophy (PhD)
Remote Sensing Image Processing
Dissertation: An Intelligent System for Land Use and Land Cover Mapping using Spaceborne Remote Sensing and GIS This study fist investigated various methods of image classification, including a Self-Organizing Map (SOM) neural network, which incorporates GIS data layer as additional input vector for segmentation and classification of a set of multi-temporal RADARSAT images for land-use and land-cover mapping. Then it investigated Multiple Classifier Combination (MCC) techniques and proposed a new method to improve classification accuracy by harnessing the goodness of the constituent classifiers. Finally, the study designed and implemented a prototype of an intelligent image processing and classification system that integrates the CLIPS expert system shell, the IDRISI Kilimanjaro GIS and image-processing software, the domain experts’ knowledge, and the USGS Geospatial Metadata via a control agent written in Visual C++.Jahangirnagar University, Bangladesh
B.Sc. (Honors) and M.Sc.
Geography
M.Sc. Research: “Flood Risk Mapping: a case study in Bansi flood plain in Dhamrai, This study investigated the socio-economical and agricultural risk level at various depth of flooding conditions and proposed a procedure of flood risk zoning based on a case study through a questionnaire survey and field measurements in “Bonshi River Flood Plain” in Bangladesh.
Publications
A Spatial Model in Development in Bangladesh
GIS ASIA/PACIFIC, Vol. 1, No. 3, Singapore
The study investigated the options of using Geographical Information Systems (GIS) in the planning and monitoring process of rural development for the improvement of roads, the methods used both spatial and attribute data. The method developed can be replicated for all the thanas (administrative area) in the country and it will not only provide a management decision tool at the micro level but will also create a composite geo-referenced database for other applications.
A Spatial Model in Development in Bangladesh
GIS ASIA/PACIFIC, Vol. 1, No. 3, Singapore
The study investigated the options of using Geographical Information Systems (GIS) in the planning and monitoring process of rural development for the improvement of roads, the methods used both spatial and attribute data. The method developed can be replicated for all the thanas (administrative area) in the country and it will not only provide a management decision tool at the micro level but will also create a composite geo-referenced database for other applications.
Implementation of Algorithms of Geocoding and SST Estimation using AVHRR Data in the EASI/PACE PCI Image Processing Software Over the Bay of Bengal.
The Journal of NOAMI, National Oceanographic and Maritime Institute, Dhaka, Bangladesh. Vol. 15, No. 2, ISSN 1027-2119
Some automatic processing modules have been developed for obtaining sea surface temperature from NOAA AVHRR data over the Bay of Bengal. The processing modules consist of geocoding, cloud detection and masking and sea surface temperature (SST) extraction. For geocoding of the raw data an orbital model has been used followed by Earth location determination algorithms. Two cloud masking techniques namely, i) gross cloud check; and ii) channel difference method have been adopted and implemented over the Bay of Bengal. Multichannel sea surface temperature (MCSST) method has been implemented over the Bay of Bengal. The accuracy of the geocoding is within 2 to 10 km and the standard deviation for sea surface temperature is within 1.0 degree Celsius. All the works have been carried out by customization EASI/PACE image processing software and the developed techniques are fully automatic. These models could be used for extracting oceanographic features from space.
A Spatial Model in Development in Bangladesh
GIS ASIA/PACIFIC, Vol. 1, No. 3, Singapore
The study investigated the options of using Geographical Information Systems (GIS) in the planning and monitoring process of rural development for the improvement of roads, the methods used both spatial and attribute data. The method developed can be replicated for all the thanas (administrative area) in the country and it will not only provide a management decision tool at the micro level but will also create a composite geo-referenced database for other applications.
Implementation of Algorithms of Geocoding and SST Estimation using AVHRR Data in the EASI/PACE PCI Image Processing Software Over the Bay of Bengal.
The Journal of NOAMI, National Oceanographic and Maritime Institute, Dhaka, Bangladesh. Vol. 15, No. 2, ISSN 1027-2119
Some automatic processing modules have been developed for obtaining sea surface temperature from NOAA AVHRR data over the Bay of Bengal. The processing modules consist of geocoding, cloud detection and masking and sea surface temperature (SST) extraction. For geocoding of the raw data an orbital model has been used followed by Earth location determination algorithms. Two cloud masking techniques namely, i) gross cloud check; and ii) channel difference method have been adopted and implemented over the Bay of Bengal. Multichannel sea surface temperature (MCSST) method has been implemented over the Bay of Bengal. The accuracy of the geocoding is within 2 to 10 km and the standard deviation for sea surface temperature is within 1.0 degree Celsius. All the works have been carried out by customization EASI/PACE image processing software and the developed techniques are fully automatic. These models could be used for extracting oceanographic features from space.
Integration of geographic information system and RADARSAT synthetic aperture radar data using a self-organizing map network as compensation for real-time ground data in automatic image classification
Journal of Applied Remote Sensing, Vol. 4, 043534
The paper presents results of using advanced techniques such as Self-Organizing feature Map (SOM) to incorporate a GIS data layer to compensate for the limited amount of real-time ground-truth data available for land-use and land-cover mapping in wet-season conditions in Bangladesh based on multi-temporal RADARSAT-1 SAR images. The experimental results were compared with those of traditional statistical classifiers such as Maximum Likelihood, Mahalanobis Distance, and Minimum Distance, which are not suitable for incorporating low-level GIS data in the image classification process. The performances of the classifiers were evaluated in terms of the classification accuracy with respect to the collected real-time ground truth data. The SOM neural network provided the highest overall accuracy when a GIS layer of land type classification with respect to the depth and duration of regular flooding was used in the network. Using this method, the overall accuracy was around 15% higher than the previously mentioned traditional classifiers at 79.6% where the training data covered only 0.53% of the total image. It also achieved higher accuracies for more classes in comparison to the other classifiers. © 2010 Society of Photo-Optical Instrumentation Engineers
A Spatial Model in Development in Bangladesh
GIS ASIA/PACIFIC, Vol. 1, No. 3, Singapore
The study investigated the options of using Geographical Information Systems (GIS) in the planning and monitoring process of rural development for the improvement of roads, the methods used both spatial and attribute data. The method developed can be replicated for all the thanas (administrative area) in the country and it will not only provide a management decision tool at the micro level but will also create a composite geo-referenced database for other applications.
Implementation of Algorithms of Geocoding and SST Estimation using AVHRR Data in the EASI/PACE PCI Image Processing Software Over the Bay of Bengal.
The Journal of NOAMI, National Oceanographic and Maritime Institute, Dhaka, Bangladesh. Vol. 15, No. 2, ISSN 1027-2119
Some automatic processing modules have been developed for obtaining sea surface temperature from NOAA AVHRR data over the Bay of Bengal. The processing modules consist of geocoding, cloud detection and masking and sea surface temperature (SST) extraction. For geocoding of the raw data an orbital model has been used followed by Earth location determination algorithms. Two cloud masking techniques namely, i) gross cloud check; and ii) channel difference method have been adopted and implemented over the Bay of Bengal. Multichannel sea surface temperature (MCSST) method has been implemented over the Bay of Bengal. The accuracy of the geocoding is within 2 to 10 km and the standard deviation for sea surface temperature is within 1.0 degree Celsius. All the works have been carried out by customization EASI/PACE image processing software and the developed techniques are fully automatic. These models could be used for extracting oceanographic features from space.
Integration of geographic information system and RADARSAT synthetic aperture radar data using a self-organizing map network as compensation for real-time ground data in automatic image classification
Journal of Applied Remote Sensing, Vol. 4, 043534
The paper presents results of using advanced techniques such as Self-Organizing feature Map (SOM) to incorporate a GIS data layer to compensate for the limited amount of real-time ground-truth data available for land-use and land-cover mapping in wet-season conditions in Bangladesh based on multi-temporal RADARSAT-1 SAR images. The experimental results were compared with those of traditional statistical classifiers such as Maximum Likelihood, Mahalanobis Distance, and Minimum Distance, which are not suitable for incorporating low-level GIS data in the image classification process. The performances of the classifiers were evaluated in terms of the classification accuracy with respect to the collected real-time ground truth data. The SOM neural network provided the highest overall accuracy when a GIS layer of land type classification with respect to the depth and duration of regular flooding was used in the network. Using this method, the overall accuracy was around 15% higher than the previously mentioned traditional classifiers at 79.6% where the training data covered only 0.53% of the total image. It also achieved higher accuracies for more classes in comparison to the other classifiers. © 2010 Society of Photo-Optical Instrumentation Engineers
Remote Sensing Techniques for Fishpond Inventory
Beleids Commisie Remote Sensing (BCRS), The Netherlands
About 80% of the ponds in Bangladesh have an area that is less than 2000 square meters and perennial water supply. This project is intended to do a rigorous test of remote sensing methods to improve the identification of ponds from its surrounding feature ... ISBN:054113049
A Spatial Model in Development in Bangladesh
GIS ASIA/PACIFIC, Vol. 1, No. 3, Singapore
The study investigated the options of using Geographical Information Systems (GIS) in the planning and monitoring process of rural development for the improvement of roads, the methods used both spatial and attribute data. The method developed can be replicated for all the thanas (administrative area) in the country and it will not only provide a management decision tool at the micro level but will also create a composite geo-referenced database for other applications.
Implementation of Algorithms of Geocoding and SST Estimation using AVHRR Data in the EASI/PACE PCI Image Processing Software Over the Bay of Bengal.
The Journal of NOAMI, National Oceanographic and Maritime Institute, Dhaka, Bangladesh. Vol. 15, No. 2, ISSN 1027-2119
Some automatic processing modules have been developed for obtaining sea surface temperature from NOAA AVHRR data over the Bay of Bengal. The processing modules consist of geocoding, cloud detection and masking and sea surface temperature (SST) extraction. For geocoding of the raw data an orbital model has been used followed by Earth location determination algorithms. Two cloud masking techniques namely, i) gross cloud check; and ii) channel difference method have been adopted and implemented over the Bay of Bengal. Multichannel sea surface temperature (MCSST) method has been implemented over the Bay of Bengal. The accuracy of the geocoding is within 2 to 10 km and the standard deviation for sea surface temperature is within 1.0 degree Celsius. All the works have been carried out by customization EASI/PACE image processing software and the developed techniques are fully automatic. These models could be used for extracting oceanographic features from space.
Integration of geographic information system and RADARSAT synthetic aperture radar data using a self-organizing map network as compensation for real-time ground data in automatic image classification
Journal of Applied Remote Sensing, Vol. 4, 043534
The paper presents results of using advanced techniques such as Self-Organizing feature Map (SOM) to incorporate a GIS data layer to compensate for the limited amount of real-time ground-truth data available for land-use and land-cover mapping in wet-season conditions in Bangladesh based on multi-temporal RADARSAT-1 SAR images. The experimental results were compared with those of traditional statistical classifiers such as Maximum Likelihood, Mahalanobis Distance, and Minimum Distance, which are not suitable for incorporating low-level GIS data in the image classification process. The performances of the classifiers were evaluated in terms of the classification accuracy with respect to the collected real-time ground truth data. The SOM neural network provided the highest overall accuracy when a GIS layer of land type classification with respect to the depth and duration of regular flooding was used in the network. Using this method, the overall accuracy was around 15% higher than the previously mentioned traditional classifiers at 79.6% where the training data covered only 0.53% of the total image. It also achieved higher accuracies for more classes in comparison to the other classifiers. © 2010 Society of Photo-Optical Instrumentation Engineers
Remote Sensing Techniques for Fishpond Inventory
Beleids Commisie Remote Sensing (BCRS), The Netherlands
About 80% of the ponds in Bangladesh have an area that is less than 2000 square meters and perennial water supply. This project is intended to do a rigorous test of remote sensing methods to improve the identification of ponds from its surrounding feature ... ISBN:054113049
Intelligent Systems for Remote Sensing Image Processing and Classification – a Review and Identification of Requirements
Proceedings of the 30th Canadian Symposium on Remote Sensing, Lethbridge, Alberta, June 2009
Recent developments in the remote sensing (RS) industry have opened up a huge potential for diverse applications, which has contributed to the rapidly increasing user community in developed and developing countries; and the frequency and the volume of Image Processing and Classification work has increased simultaneously. However, the number of experts in the area has not increased at the same pace. The available data, tools, and techniques are too complex for most users, unless they devote a considerable amount of time to obtaining the relevant “technical” background. Users are faced with the problems of viewing a mass of data, applying appropriate methods, evaluating the results, and handling the specific computer platform. Therefore, not only are advanced methods required for diverse and complex image processing and classification works, but an intelligent system is required, which incorporates advanced methods and reduces the dependency on experts. In this context, this paper reviews works on intelligent systems in RS image processing and classifications and identifies current requirements for the design of an intelligent RS image processing and classification system. ”
A Spatial Model in Development in Bangladesh
GIS ASIA/PACIFIC, Vol. 1, No. 3, Singapore
The study investigated the options of using Geographical Information Systems (GIS) in the planning and monitoring process of rural development for the improvement of roads, the methods used both spatial and attribute data. The method developed can be replicated for all the thanas (administrative area) in the country and it will not only provide a management decision tool at the micro level but will also create a composite geo-referenced database for other applications.
Implementation of Algorithms of Geocoding and SST Estimation using AVHRR Data in the EASI/PACE PCI Image Processing Software Over the Bay of Bengal.
The Journal of NOAMI, National Oceanographic and Maritime Institute, Dhaka, Bangladesh. Vol. 15, No. 2, ISSN 1027-2119
Some automatic processing modules have been developed for obtaining sea surface temperature from NOAA AVHRR data over the Bay of Bengal. The processing modules consist of geocoding, cloud detection and masking and sea surface temperature (SST) extraction. For geocoding of the raw data an orbital model has been used followed by Earth location determination algorithms. Two cloud masking techniques namely, i) gross cloud check; and ii) channel difference method have been adopted and implemented over the Bay of Bengal. Multichannel sea surface temperature (MCSST) method has been implemented over the Bay of Bengal. The accuracy of the geocoding is within 2 to 10 km and the standard deviation for sea surface temperature is within 1.0 degree Celsius. All the works have been carried out by customization EASI/PACE image processing software and the developed techniques are fully automatic. These models could be used for extracting oceanographic features from space.
Integration of geographic information system and RADARSAT synthetic aperture radar data using a self-organizing map network as compensation for real-time ground data in automatic image classification
Journal of Applied Remote Sensing, Vol. 4, 043534
The paper presents results of using advanced techniques such as Self-Organizing feature Map (SOM) to incorporate a GIS data layer to compensate for the limited amount of real-time ground-truth data available for land-use and land-cover mapping in wet-season conditions in Bangladesh based on multi-temporal RADARSAT-1 SAR images. The experimental results were compared with those of traditional statistical classifiers such as Maximum Likelihood, Mahalanobis Distance, and Minimum Distance, which are not suitable for incorporating low-level GIS data in the image classification process. The performances of the classifiers were evaluated in terms of the classification accuracy with respect to the collected real-time ground truth data. The SOM neural network provided the highest overall accuracy when a GIS layer of land type classification with respect to the depth and duration of regular flooding was used in the network. Using this method, the overall accuracy was around 15% higher than the previously mentioned traditional classifiers at 79.6% where the training data covered only 0.53% of the total image. It also achieved higher accuracies for more classes in comparison to the other classifiers. © 2010 Society of Photo-Optical Instrumentation Engineers
Remote Sensing Techniques for Fishpond Inventory
Beleids Commisie Remote Sensing (BCRS), The Netherlands
About 80% of the ponds in Bangladesh have an area that is less than 2000 square meters and perennial water supply. This project is intended to do a rigorous test of remote sensing methods to improve the identification of ponds from its surrounding feature ... ISBN:054113049
Intelligent Systems for Remote Sensing Image Processing and Classification – a Review and Identification of Requirements
Proceedings of the 30th Canadian Symposium on Remote Sensing, Lethbridge, Alberta, June 2009
Recent developments in the remote sensing (RS) industry have opened up a huge potential for diverse applications, which has contributed to the rapidly increasing user community in developed and developing countries; and the frequency and the volume of Image Processing and Classification work has increased simultaneously. However, the number of experts in the area has not increased at the same pace. The available data, tools, and techniques are too complex for most users, unless they devote a considerable amount of time to obtaining the relevant “technical” background. Users are faced with the problems of viewing a mass of data, applying appropriate methods, evaluating the results, and handling the specific computer platform. Therefore, not only are advanced methods required for diverse and complex image processing and classification works, but an intelligent system is required, which incorporates advanced methods and reduces the dependency on experts. In this context, this paper reviews works on intelligent systems in RS image processing and classifications and identifies current requirements for the design of an intelligent RS image processing and classification system. ”
Identifying the Morphological Changes of a Distributary of the Ganges in Response to the Declining Flow Using Remote Sensing
Proceedings of the 20th Asian Conference on Remote Sensing, Vol. 1
Cartographic evidence (Rennell, 1776) shows that the Gorai River has been a major distributary of the Ganges river system since at least 200 years ago. It was a perennial river until the end of the last decade. The Gorai River carries almost 8 to 12% of the discharge of the Ganges River. This river is the main source of fresh water inflow into the southwest region of Bangladesh. The importance of this river varies with time and is governed mainly by the planform of its off-take with the Ganges River. The Farakka Barrage inside India, 18 km upstream of the international border, became operational in 1975. Since then the barrage has been diverting a substantial amount of dry season flow of the Ganges River through the Hoogly River and has thus initiated an irreversible process of deterioration of the Gorai River flow.
