Akansha Mehrotra | Indian Institute Of Technology, Roorkee (original) (raw)

Papers by Akansha Mehrotra

students conference on engineering and systems, Apr 12, 2013

ABSTRACT This paper presents a new technique for unsupervised change detection in bitemporal remo... more ABSTRACT This paper presents a new technique for unsupervised change detection in bitemporal remote sensing images using spectral change difference images and hybrid genetic FCM. The proposed method works in three steps. In the first step, three spectral change difference images:absolute value difference image, ratio image and log ratio image are computed. In the next step, a feature vector space is created using PCA. Finally, the change detection is obtained by dividing the feature vector space into two clusters using genetic FCM. The validity of the clusters is measured by DB index. The parts of image of Reno-Lake Tahoe area was used as data set for the performance evaluation of proposed algorithm. The results obtained were compared with EM based, MRF based and NSCT methods. The results verify that the proposed algorithm provides superior results than the other existing methods.

This paper presents a new technique for unsupervised change detection in bitemporal remote sensin... more This paper presents a new technique for unsupervised change detection in bitemporal remote sensing images using spectral change difference images and hybrid genetic FCM. The proposed method works in three steps. In the first step, three spectral change difference images:absolute value difference image, ratio image and log ratio image are computed. In the next step, a feature vector space is created using PCA. Finally, the change detection is obtained by dividing the feature vector space into two clusters using genetic FCM. The validity of the clusters is measured by DB index. The parts of image of Reno-Lake Tahoe area was used as data set for the performance evaluation of proposed algorithm. The results obtained were compared with EM based, MRF based and NSCT methods. The results verify that the proposed algorithm provides superior results than the other existing methods.

Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering, 2013

IJCSI International Journal of Computer Science Issues, Sep 1, 2010

In computer vision, segmentation refers to the process of partitioning a digital image into multi... more In computer vision, segmentation refers to the process of partitioning a digital image into multiple segments (sets of pixels, also known as superpixels).Image segmentation is typically used to locate objects and boundaries (lines, curves, etc.) in images. More precisely, image ...

IJCSI, 2010

IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 3, No 11, May 2010 ISSN (On... more IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 3, No 11, May 2010 ISSN (Online): 1694-0784 ISSN (Print): 1694-0814 28 Faster and Efficient Web Crawling with Parallel Migrating Web Crawler Akansha Singh1, Krishna Kant Singh2 1Deptt. Of Information ...

International Journal of Computer Applications, 2012

International Journal of Advanced Research in Electrical, Electronics and Instrumentation Energy, 2014

This paper demonstrates a CMOS frequency synthesizer design, whose primary purpose is to test the... more This paper demonstrates a CMOS frequency synthesizer design, whose primary purpose is to test the designer’s high speed, mixed-signal CMOS circuit design skill. Line codes are the techniques for representing digital sequences by pulse waveforms suitable for baseband transmission. NRZ or non return to zero is an important Line coding method. NRZ pulses are of full bit duration. We do not get square waveform in conventional current starved VCO. So, the conventional current starved VCO cannot be used for generating NRZ line coding as it is necessary that the output waveform of VCO should be square wave for NRZ coding. This new current starved CMOS VCO is used to design a DPLL. Furthermore, this DPLL design is used to generate a clock for a 8.33Mbits/second with NRZ data format for center frequency at VCO. The DPLL presented here uses XOR phase detector for reducing jitter noise and divide by two stage is used in the feedback loop for frequency synthesis. The DPLL is designed uses activ...

International Journal of Computer Applications, 2012

detection is one of the most important tasks in the field of image processing. Detection of edges... more detection is one of the most important tasks in the field of image processing. Detection of edges from noisy images is of greater importance as most images obtained are corrupted by impulse noise due to communication and transmission errors. In the proposed work a novel adaptive algorithm for finding edges of noisy images is proposed. One of the major problem with noisy images is that the noise pixels are also detected as edges, so in the proposed algorithm a threshold is used to distinguish between edge pixels and noise pixels. The value of threshold is inversely proportional to the level of details whose edges are detected and makes the algorithm adaptive. A sliding window is taken and the difference of the center pixel with all the pixels of the window whose value is not equal to zero or one is taken, But if the center pixel itself is zero or one then the difference from a noise free median is calculated. The average of the differences is checked against a threshold value. If the...

Priyanka Khandelwal , Krishna Kant Singh , B.K.Singh , Akansha Mehrotra 1 Electronics & Communica... more Priyanka Khandelwal , Krishna Kant Singh , B.K.Singh , Akansha Mehrotra 1 Electronics & Communication Engineering, BTKIT, Dwarhat, India 2 Earthquake Engineering, IIT, Roorkee, India 1 p.khandelwal.ece@gmail.com bksapkec@yahoo.com 2 krishnaiitr2011@gmail.com akanshasing@gmail.com Abstract— In this paper, an unsupervised change detection technique of multispectral images based on wavelet fusion and Kohonen Clustering Network is presented. The proposed method fuses absolute difference and change vector analysis image using wavelet fusion rules. The fused image highlights the changed areas while suppress unchanged areas. The Kohonen clustering network is used to create the final change map with changed areas highlighted. The performance of the proposed method was tested on two Landsat 5 TM images of Alaska region. The results obtained were compared with some other existing state of the art methods and it is observed that the proposed method outperforms the other methods. KeywordCVA, Wa...

This paper presents an automatic method for exudate detection from colour fundus images based on ... more This paper presents an automatic method for exudate detection from colour fundus images based on Differential Morphological Profile (DMP).The detection of exudates is important for the identification of eye diseases such as diabetic retinopathy. The method involves of three main phases. In the first phase, pre processing tasks like Gaussian smoothing and contrast enhancement is done. In the second phase, DMP is applied on the pre-processed image. The image obtained from DMP contains highlighted bright regions consisting of exudates and optic disc. In the next phase, feature extraction based on location of optic disc, shape index and area is done to obtain actual exudates. The performance of the proposed method is evaluated by applying it on the DIARETDB1 database. The specificity, sensitivity and PPV of the proposed method were compared with two other methods. The results show that the proposed method gives better results than the other conventional methods. Keyword- DMP, Exudates, ...

2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS

2014 Ieee International Advance Computing Conference, Feb 1, 2014

ABSTRACT This paper presents an automated blood vessel detection method from the fundus image. Th... more ABSTRACT This paper presents an automated blood vessel detection method from the fundus image. The method first performs some basic image preprocessing tasks on the green channel of the retinal image. A combination of morphological operations like top- hat and bottom-hat transformations are applied on the preprocessed image to highlight the blood vessels. Finally, the Kohonen Clustering Network is applied to cluster the input image into two clusters namely vessel and non-vessel. The performance of the proposed method is tested by applying it on retinal images from Digital Retinal Images for Vessel Extraction (DRIVE)database. The results obtained from the proposed method are compared with three other state of the art methods. The sensitivity, false-positive fraction (FPF) and accuracy of the proposed method is found to be higher than the other methods which imply that the proposed method is more efficient and accurate.

IETE Technical Review, 2014

2014 International Conference on Computing for Sustainable Global Development (INDIACom), 2014

An automatic building extraction method for very high resolution satellite images is presented in... more An automatic building extraction method for very high resolution satellite images is presented in this paper. Two novel operators Differential Morphological Shadow Operator (DMSO) and Differential Morphological Building Operator (DMBO) for extracting shadows and buildings respectively are proposed. The method identifies shadows using DMSO. After masking out the shadow, DMBO is applied to obtain the building candidates. The building candidates are filtered using four parameters including shadow test, Shape Index (SI), area and rectangular fit to remove false detections and obtain only buildings. The proposed method was tested on two VHR satellite images acquired by Geoeye-1.The results of the proposed method were compared with two other state of the art methods and it was observed that the proposed method outperformed the other two methods.

Natural Hazards, 2015

ABSTRACT The coastal areas of Japan were hard hit by a magnitude 9.0 earthquake on 11 March 2011.... more ABSTRACT The coastal areas of Japan were hard hit by a magnitude 9.0 earthquake on 11 March 2011. The earthquake triggered a disastrous tsunami over the area which led to massive destruction. In this paper, tsunami-induced changes in Soma, Watari, Natori and Iwanuma areas using Landsat 7 ETM+ and EO-1 ALI images are identified. The proposed method is based on image classification using radial basis function neural network and generalized improved fuzzy partition FCM algorithm. The pre- and post-tsunami images of the area are first classified using a radial basis function neural network. The pre- and post-tsunami images are classified into three classes including water, vegetation and urban and bare land class. The classified images are compared with other to obtain a set of four change classes. These change classes are labelled to obtain a classified change map. The change map reveals that large areas of vegetations and urban land are washed away by the tsunami in all the four cities, Soma, Watari, Natori and Iwanuma. The accuracy assessment of the method shows that the results obtained are quite satisfactory. The method has high overall accuracy and kappa coefficient value.

Natural Hazards, 2015

ABSTRACT The coastal areas of Japan were hard hit by a magnitude 9.0 earthquake on 11 March 2011.... more ABSTRACT The coastal areas of Japan were hard hit by a magnitude 9.0 earthquake on 11 March 2011. The earthquake triggered a disastrous tsunami over the area which led to massive destruction. In this paper, tsunami-induced changes in Soma, Watari, Natori and Iwanuma areas using Landsat 7 ETM+ and EO-1 ALI images are identified. The proposed method is based on image classification using radial basis function neural network and generalized improved fuzzy partition FCM algorithm. The pre- and post-tsunami images of the area are first classified using a radial basis function neural network. The pre- and post-tsunami images are classified into three classes including water, vegetation and urban and bare land class. The classified images are compared with other to obtain a set of four change classes. These change classes are labelled to obtain a classified change map. The change map reveals that large areas of vegetations and urban land are washed away by the tsunami in all the four cities, Soma, Watari, Natori and Iwanuma. The accuracy assessment of the method shows that the results obtained are quite satisfactory. The method has high overall accuracy and kappa coefficient value.

ISET GOLDEN JUBILEE SYMPOSIUM

This paper presents a supervised change detection technique for satellite images using a probabil... more This paper presents a supervised change detection technique for satellite images using a probabilistic neural network (PNN). The proposed method works in two phases. In the first phase a difference image is computed. The most commonly used techniques for computing the difference image such as ratio images or log ratio images degrade the performance of the algorithm in the presence of speckle noise. To overcome the above mentioned limitations the difference image in this work is computed using normalized neighborhood ratio based method. In the next phase the PNN is used to detect efficiently any change between the two images. An estimator is used by the PNN to estimate the probability density function. The ratio of two conditional probability density functions, called the likelihood ratio is computed. Finally, the log likelihood ratio test is used to classify the pixels of the difference image into changed and unchanged classes to create a change map. The change map highlights the changes that have occurred between the two input images. The proposed method was compared quantatively as well as qualitatively with other existing state of the art methods. The results showed that the proposed method outperforms the other methods.

students conference on engineering and systems, Apr 12, 2013

ABSTRACT This paper presents a new technique for unsupervised change detection in bitemporal remo... more ABSTRACT This paper presents a new technique for unsupervised change detection in bitemporal remote sensing images using spectral change difference images and hybrid genetic FCM. The proposed method works in three steps. In the first step, three spectral change difference images:absolute value difference image, ratio image and log ratio image are computed. In the next step, a feature vector space is created using PCA. Finally, the change detection is obtained by dividing the feature vector space into two clusters using genetic FCM. The validity of the clusters is measured by DB index. The parts of image of Reno-Lake Tahoe area was used as data set for the performance evaluation of proposed algorithm. The results obtained were compared with EM based, MRF based and NSCT methods. The results verify that the proposed algorithm provides superior results than the other existing methods.

This paper presents a new technique for unsupervised change detection in bitemporal remote sensin... more This paper presents a new technique for unsupervised change detection in bitemporal remote sensing images using spectral change difference images and hybrid genetic FCM. The proposed method works in three steps. In the first step, three spectral change difference images:absolute value difference image, ratio image and log ratio image are computed. In the next step, a feature vector space is created using PCA. Finally, the change detection is obtained by dividing the feature vector space into two clusters using genetic FCM. The validity of the clusters is measured by DB index. The parts of image of Reno-Lake Tahoe area was used as data set for the performance evaluation of proposed algorithm. The results obtained were compared with EM based, MRF based and NSCT methods. The results verify that the proposed algorithm provides superior results than the other existing methods.

Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering, 2013

IJCSI International Journal of Computer Science Issues, Sep 1, 2010

In computer vision, segmentation refers to the process of partitioning a digital image into multi... more In computer vision, segmentation refers to the process of partitioning a digital image into multiple segments (sets of pixels, also known as superpixels).Image segmentation is typically used to locate objects and boundaries (lines, curves, etc.) in images. More precisely, image ...

IJCSI, 2010

IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 3, No 11, May 2010 ISSN (On... more IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 3, No 11, May 2010 ISSN (Online): 1694-0784 ISSN (Print): 1694-0814 28 Faster and Efficient Web Crawling with Parallel Migrating Web Crawler Akansha Singh1, Krishna Kant Singh2 1Deptt. Of Information ...

International Journal of Computer Applications, 2012

International Journal of Advanced Research in Electrical, Electronics and Instrumentation Energy, 2014

This paper demonstrates a CMOS frequency synthesizer design, whose primary purpose is to test the... more This paper demonstrates a CMOS frequency synthesizer design, whose primary purpose is to test the designer’s high speed, mixed-signal CMOS circuit design skill. Line codes are the techniques for representing digital sequences by pulse waveforms suitable for baseband transmission. NRZ or non return to zero is an important Line coding method. NRZ pulses are of full bit duration. We do not get square waveform in conventional current starved VCO. So, the conventional current starved VCO cannot be used for generating NRZ line coding as it is necessary that the output waveform of VCO should be square wave for NRZ coding. This new current starved CMOS VCO is used to design a DPLL. Furthermore, this DPLL design is used to generate a clock for a 8.33Mbits/second with NRZ data format for center frequency at VCO. The DPLL presented here uses XOR phase detector for reducing jitter noise and divide by two stage is used in the feedback loop for frequency synthesis. The DPLL is designed uses activ...

International Journal of Computer Applications, 2012

detection is one of the most important tasks in the field of image processing. Detection of edges... more detection is one of the most important tasks in the field of image processing. Detection of edges from noisy images is of greater importance as most images obtained are corrupted by impulse noise due to communication and transmission errors. In the proposed work a novel adaptive algorithm for finding edges of noisy images is proposed. One of the major problem with noisy images is that the noise pixels are also detected as edges, so in the proposed algorithm a threshold is used to distinguish between edge pixels and noise pixels. The value of threshold is inversely proportional to the level of details whose edges are detected and makes the algorithm adaptive. A sliding window is taken and the difference of the center pixel with all the pixels of the window whose value is not equal to zero or one is taken, But if the center pixel itself is zero or one then the difference from a noise free median is calculated. The average of the differences is checked against a threshold value. If the...

Priyanka Khandelwal , Krishna Kant Singh , B.K.Singh , Akansha Mehrotra 1 Electronics & Communica... more Priyanka Khandelwal , Krishna Kant Singh , B.K.Singh , Akansha Mehrotra 1 Electronics & Communication Engineering, BTKIT, Dwarhat, India 2 Earthquake Engineering, IIT, Roorkee, India 1 p.khandelwal.ece@gmail.com bksapkec@yahoo.com 2 krishnaiitr2011@gmail.com akanshasing@gmail.com Abstract— In this paper, an unsupervised change detection technique of multispectral images based on wavelet fusion and Kohonen Clustering Network is presented. The proposed method fuses absolute difference and change vector analysis image using wavelet fusion rules. The fused image highlights the changed areas while suppress unchanged areas. The Kohonen clustering network is used to create the final change map with changed areas highlighted. The performance of the proposed method was tested on two Landsat 5 TM images of Alaska region. The results obtained were compared with some other existing state of the art methods and it is observed that the proposed method outperforms the other methods. KeywordCVA, Wa...

This paper presents an automatic method for exudate detection from colour fundus images based on ... more This paper presents an automatic method for exudate detection from colour fundus images based on Differential Morphological Profile (DMP).The detection of exudates is important for the identification of eye diseases such as diabetic retinopathy. The method involves of three main phases. In the first phase, pre processing tasks like Gaussian smoothing and contrast enhancement is done. In the second phase, DMP is applied on the pre-processed image. The image obtained from DMP contains highlighted bright regions consisting of exudates and optic disc. In the next phase, feature extraction based on location of optic disc, shape index and area is done to obtain actual exudates. The performance of the proposed method is evaluated by applying it on the DIARETDB1 database. The specificity, sensitivity and PPV of the proposed method were compared with two other methods. The results show that the proposed method gives better results than the other conventional methods. Keyword- DMP, Exudates, ...

2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS

2014 Ieee International Advance Computing Conference, Feb 1, 2014

ABSTRACT This paper presents an automated blood vessel detection method from the fundus image. Th... more ABSTRACT This paper presents an automated blood vessel detection method from the fundus image. The method first performs some basic image preprocessing tasks on the green channel of the retinal image. A combination of morphological operations like top- hat and bottom-hat transformations are applied on the preprocessed image to highlight the blood vessels. Finally, the Kohonen Clustering Network is applied to cluster the input image into two clusters namely vessel and non-vessel. The performance of the proposed method is tested by applying it on retinal images from Digital Retinal Images for Vessel Extraction (DRIVE)database. The results obtained from the proposed method are compared with three other state of the art methods. The sensitivity, false-positive fraction (FPF) and accuracy of the proposed method is found to be higher than the other methods which imply that the proposed method is more efficient and accurate.

IETE Technical Review, 2014

2014 International Conference on Computing for Sustainable Global Development (INDIACom), 2014

An automatic building extraction method for very high resolution satellite images is presented in... more An automatic building extraction method for very high resolution satellite images is presented in this paper. Two novel operators Differential Morphological Shadow Operator (DMSO) and Differential Morphological Building Operator (DMBO) for extracting shadows and buildings respectively are proposed. The method identifies shadows using DMSO. After masking out the shadow, DMBO is applied to obtain the building candidates. The building candidates are filtered using four parameters including shadow test, Shape Index (SI), area and rectangular fit to remove false detections and obtain only buildings. The proposed method was tested on two VHR satellite images acquired by Geoeye-1.The results of the proposed method were compared with two other state of the art methods and it was observed that the proposed method outperformed the other two methods.

Natural Hazards, 2015

ABSTRACT The coastal areas of Japan were hard hit by a magnitude 9.0 earthquake on 11 March 2011.... more ABSTRACT The coastal areas of Japan were hard hit by a magnitude 9.0 earthquake on 11 March 2011. The earthquake triggered a disastrous tsunami over the area which led to massive destruction. In this paper, tsunami-induced changes in Soma, Watari, Natori and Iwanuma areas using Landsat 7 ETM+ and EO-1 ALI images are identified. The proposed method is based on image classification using radial basis function neural network and generalized improved fuzzy partition FCM algorithm. The pre- and post-tsunami images of the area are first classified using a radial basis function neural network. The pre- and post-tsunami images are classified into three classes including water, vegetation and urban and bare land class. The classified images are compared with other to obtain a set of four change classes. These change classes are labelled to obtain a classified change map. The change map reveals that large areas of vegetations and urban land are washed away by the tsunami in all the four cities, Soma, Watari, Natori and Iwanuma. The accuracy assessment of the method shows that the results obtained are quite satisfactory. The method has high overall accuracy and kappa coefficient value.

Natural Hazards, 2015

ABSTRACT The coastal areas of Japan were hard hit by a magnitude 9.0 earthquake on 11 March 2011.... more ABSTRACT The coastal areas of Japan were hard hit by a magnitude 9.0 earthquake on 11 March 2011. The earthquake triggered a disastrous tsunami over the area which led to massive destruction. In this paper, tsunami-induced changes in Soma, Watari, Natori and Iwanuma areas using Landsat 7 ETM+ and EO-1 ALI images are identified. The proposed method is based on image classification using radial basis function neural network and generalized improved fuzzy partition FCM algorithm. The pre- and post-tsunami images of the area are first classified using a radial basis function neural network. The pre- and post-tsunami images are classified into three classes including water, vegetation and urban and bare land class. The classified images are compared with other to obtain a set of four change classes. These change classes are labelled to obtain a classified change map. The change map reveals that large areas of vegetations and urban land are washed away by the tsunami in all the four cities, Soma, Watari, Natori and Iwanuma. The accuracy assessment of the method shows that the results obtained are quite satisfactory. The method has high overall accuracy and kappa coefficient value.

ISET GOLDEN JUBILEE SYMPOSIUM

This paper presents a supervised change detection technique for satellite images using a probabil... more This paper presents a supervised change detection technique for satellite images using a probabilistic neural network (PNN). The proposed method works in two phases. In the first phase a difference image is computed. The most commonly used techniques for computing the difference image such as ratio images or log ratio images degrade the performance of the algorithm in the presence of speckle noise. To overcome the above mentioned limitations the difference image in this work is computed using normalized neighborhood ratio based method. In the next phase the PNN is used to detect efficiently any change between the two images. An estimator is used by the PNN to estimate the probability density function. The ratio of two conditional probability density functions, called the likelihood ratio is computed. Finally, the log likelihood ratio test is used to classify the pixels of the difference image into changed and unchanged classes to create a change map. The change map highlights the changes that have occurred between the two input images. The proposed method was compared quantatively as well as qualitatively with other existing state of the art methods. The results showed that the proposed method outperforms the other methods.