The reinforced two layered foundation bed considered for study consists of a layer of granular fill overlying soft non-homogeneous clay with inclusion or reinforcement （geosymhetic strips, grids or sheets） in single...The reinforced two layered foundation bed considered for study consists of a layer of granular fill overlying soft non-homogeneous clay with inclusion or reinforcement （geosymhetic strips, grids or sheets） in single layer at soil-granular fill interface A method is developed to estimate the bearing capacity of a strip footing on the surface of a reinforced foundation bed over a finite layer of clay whose undrained strength increases linearly with depth incorporating the contribution of axial resistance of the reinforcement together with those of granular fill and soft ground. Parametric studies presented quantify the improvement in bearing capacity.展开更多
It is common knowledge that the end user of stream flow data may necessarily not have any prior knowledge of the quality control measures applied in their generation, therefore, conclusions drawn most often times may ...It is common knowledge that the end user of stream flow data may necessarily not have any prior knowledge of the quality control measures applied in their generation, therefore, conclusions drawn most often times may not be effective as desired. Thus, this study is an attempt at providing an independent quality construct to boost the confidence in the use of stream flow data by developing regional flow duration curves for selected ungauged stations of the upper Niger River Basin, Nigeria. Toward this end, stream flow data for seven gauging stations cover some sub basins in the Basin were obtained;precisely, monthly stream flow data covering a range of eleven to fifty-three years period. The flow duration curves from the gauging stations were fitted with three probability distribution models;i.e., logarithmic, power and exponential regression models. For the regionalisation, parameterisation was carried out in terms of the drainage area alone to allow for simplicity of models. Results obtained showed that the exponential regression model, in terms of Coefficient of Determination (R<sup>2</sup>) had the best fit. Though the regionalised model was simple, measurable agreement was obtained during the calibration and validation phases. However, considering the length of data used and probable variability in the stream flow regime, it is not possible to objectively generalise on the quality of the results. Against this backdrop, it suffices to take into cognisance the need to use an ensemble of catchment characteristics in the development of the flow duration curves and the overall regional models;this is important considering the implications of anthropogenic activities and hydro-climatic variations.展开更多
India is an agricultural country and a core source of income for the world population.The Indian economy is greatly depending on agriculture that is decrease day by day due to pandemic COVID-19.India is a major export...India is an agricultural country and a core source of income for the world population.The Indian economy is greatly depending on agriculture that is decrease day by day due to pandemic COVID-19.India is a major exporter of many crop foods.India,Thailand,and Vietnam are the major exports of rice if these stopped exports it reduces the economy up to 15%.A related circumstance is built up with diverse yields too like wheat,sunflower whose fare has been stationary by Kazakhstan,Serbia individually.In India,the end of April is the main source of income to farmers because they sell their rabi crops(wheat,mustard,maize,lentil,chilies,gram,tomatoes)in the market drastically decreases of CFPI may lead to the distress of Indian agricultural economy.The change over time in the price of options on wheat futures reveals increased price volatility in response to growing uncertainty about the COVID-19 impacts.展开更多
Internet of Things(IoT)is a recent paradigm to improve human lifestyle.Nowadays,number devices are connected to the Internet drastically.Thus,the people can control and monitor the physical things in real-time without...Internet of Things(IoT)is a recent paradigm to improve human lifestyle.Nowadays,number devices are connected to the Internet drastically.Thus,the people can control and monitor the physical things in real-time without delay.The IoT plays a vital role in all kind of fields in our world such as agriculture,livestock,transport,and healthcare,grid system,connected home,elderly people carrying system,cypher physical system,retail,and intelligent systems.In IoT energy conservation is a challenging task,as the devices are made up of low-cost and low-power sensing devices and local processing.IoT networks have significant challenges in two areas:network lifespan and energy usage.Therefore,the clustering is a right choice to prolong the energy in the network.In LEACH clustering protocol,sometimes the same node acts as CH again and again probabilistically.To overcome these issues,this paper proposes the Energy-Aware Cluster-based Routing(EACRLEACH)protocol in WSN based IoT.The Cluster Head(CH)selection is a crucial task in clustering protocol inWSN based IoT.In EACR-LEACH,the CH is selected by using the routing metrics,Residual Energy(RER),Number of Neighbors(NoN),Distance between Sensor Node and Sink(Distance)and Number of Time Node Act as CH(NTNACH).An extensive simulation is conducted on MATLAB 2019a.The accomplishment of EACR-LEACH is compared to LEACH and SE-LEACH.The proposed EACR-LEACH protocol extends the network’s lifetime by 4%-8%and boosts throughput by 16%–24%.展开更多
In the present scenario,Deep Learning(DL)is one of the most popular research algorithms to increase the accuracy of data analysis.Due to intra-class differences and inter-class variation,image classification is one of...In the present scenario,Deep Learning(DL)is one of the most popular research algorithms to increase the accuracy of data analysis.Due to intra-class differences and inter-class variation,image classification is one of the most difficult jobs in image processing.Plant or spinach recognition or classification is one of the deep learning applications through its leaf.Spinach is more critical for human skin,bone,and hair,etc.It provides vitamins,iron,minerals,and protein.It is beneficial for diet and is readily available in people’s surroundings.Many researchers have proposed various machine learning and deep learning algorithms to classify plant images more accurately in recent years.This paper presents a novel Convolutional Neural Network(CNN)to recognize spinach more accurately.The proposed CNN architecture classifies the spinach category,namely Amaranth leaves,Black nightshade,Curry leaves,and Drumstick leaves.The dataset contains 400 images with four classes,and each type has 100 images.The images were captured from the agricultural land located at Thirumanur,Salem district,Tamil Nadu.The proposed CNN achieves 97.5%classification accuracy.In addition,the performance of the proposed CNN is compared with Support Vector Machine(SVM),Random Forest,Visual Geometry Group 16(VGG16),Visual Geometry Group 19(VGG19)and Residual Network 50(ResNet50).The proposed provides superior performance than other models,namely SVM,Random Forest,VGG16,VGG19 and ResNet50.展开更多
Food security for the 7 billion people on earth requires minimizing crop damage by timely detectionofdiseases.Most deep learningmodels forautomated detectionof diseases in plants suffer fromthe fatal flaw that once te...Food security for the 7 billion people on earth requires minimizing crop damage by timely detectionofdiseases.Most deep learningmodels forautomated detectionof diseases in plants suffer fromthe fatal flaw that once tested on independent data,their performance drops significantly.This work investigates a potential solution to this problem by using segmented image data to train the convolutional neural network(CNN)models.As compared to the F-CNN model trained using full images,S-CNN model trained using segmented imagesmore than doubles in performance to 98.6%accuracy when tested on independent data previously unseen by the models even with 10 disease classes.Not only this,by using tomato plant and target spot disease type as an example,we show that the confidence of self-classification for S-CNN model improves significantly over F-CNN model.This research work brings applicability of automated methods closer to non-experts for timely detection of diseases.展开更多
Unbalanced operating condition in a power system can cause partial overloading of the generators in the network,a condition where one or two of the three phases of the generator become overloaded even if the total 3-p...Unbalanced operating condition in a power system can cause partial overloading of the generators in the network,a condition where one or two of the three phases of the generator become overloaded even if the total 3-phase power output of the generator is within its specified limit.Partial overloading of generators beyond certain limits is undesirable and must be avoided.Distribution systems are often subjected to highly unbalanced operating conditions.Introduction of distributed generations(DGs),therefore,has rendered today’s distribution systems quite susceptible to this problem.Mitigation of this problem requires the issue to be addressed properly during analysis,operation and planning of such systems.Analysis,operation and planning of power networks under unbalanced operating condition require 3-phase load flow study.The existing methods of 3-phase load flow are not equipped to take into account any limit on the loadings of the individual phases of the generators.In the present work,a methodology based on NewtonRaphson(N-R)3-phase load flow with necessary modifications is proposed.The proposed methodology is able to determine the safe loading limits of the generators,and,can be adopted for operation and planning of power networks under unbalanced operating conditions to overcome the above difficulties.Test results on IEEE-37 bus feeder network are presented to demonstrate the effectiveness of the proposed method.展开更多
In the field of digital image processing,denoising is one of the basic problems.The challenges faced in image denoising are detecting impulse noise and designing a suitable filter.In this paper,we propose a methodolog...In the field of digital image processing,denoising is one of the basic problems.The challenges faced in image denoising are detecting impulse noise and designing a suitable filter.In this paper,we propose a methodology to remove the random impulse noise on the color image using a novel switching median filter.By using this novel technique,the occurrence of color artifacts has been avoided after noise removal which depends on auto-tuning threshold detection and a vector-type median filter noise remover.In the proposed technique,the random valued impulse noises with uniform distribution have been dealt with switching median filter.L_(2)Norm is employed to calculate the distribution distance rather than L1 Norm which is used to identify the optimal threshold value for auto-tuning filter.The switching auto-tuning detector automatically tunes the noisy pixels based on distance information of pixels distribution.The Normalized Mean Square Error(NMSE)is found to decrease for L_(2)Norm when compared with L1 Norm.The Peak Signal to Noise Ratio(PSNR)value and True Positive Rate(TPR)value improved with L_(2)Norm signifying effective noise removal.The efficiency of the present method is verified by conducting experiments on digital images.展开更多
The current milieu,encourages rapid growth of wireless communication,multimedia applications,robotics and graphics to have efficient utilization of resources with high throughput and low power digital signal processin...The current milieu,encourages rapid growth of wireless communication,multimedia applications,robotics and graphics to have efficient utilization of resources with high throughput and low power digital signal processing(DSP)systems.In an aggregate DSP system ranging from audio/video signal processing to wireless sensor networks,floating point matrix multiplication is used in wide scale in most of the fundamental processing units.Hardware implementation of floating-point matrix multiplication demands a colossal number of arithmetic operations that alter speed and consuming more area and power.DSP systems essentially uses two techniques to reduce dynamic power consumption:-they are pipelining and parallel processing that needs high performance processing element with less area and low power in diverse scientific computing applications.However,number of adders and multipliers used in the design of floating-point unit also increases subsequently.The adders and multipliers are the most area,delay and power consuming data path elements in the processing unit.The arithmetic level reduction of delay,power and area in the processing element is performed by the selection of appropriate adders and multipliers.This article proposes a parallel multiplication architecture using Strassen and UrdhvaTiryagbhyam multiplier,which involves design of efficient parallel matrix multiplication with flexible implementation of FPGA(Field Programmable Gate Array)device to analyse the computation and area.The design incorporates scheduling of blocks,operations on processing elements,block size determination,parallelization and double buffering for storage of matrix elements.展开更多
The adaptive sign least mean square(SLMS)filter may change dynamically depending on the filter output.Noise cancellation is one of the most common adaptive filter applications.In real-time applications such as medical...The adaptive sign least mean square(SLMS)filter may change dynamically depending on the filter output.Noise cancellation is one of the most common adaptive filter applications.In real-time applications such as medical computing,the speed of the process producing hardware is critical,hence,this study proposes the hardware implementation of the SLMS adaptive filter utilizing the Xilinx System Generator.The suggested design seeks to improve performance while decreasing convergence rate and route latency.In this paper,(i)we proposed a modified architecture for an 8-tap SLMS adaptive filter is built,and(ii)multiplier-less structure for a Modified DLMS Filter is introduced and compare the same.The ECG signal was evaluated using the intended architecture.The algorithm’s functionality is tested in MATLAB using varied ECG data from the MIT-BIH database as input.Using Xilinx system generator,the LMS and SLMS are developed,simulated,synthesized,and implemented on a Virtex-5 FPGA.When comparing Systolic Sign LMS Filter to LMS Filter,the result reveals a 5%drop in total real-time router completion and a drop in the number of adders and subtractors,as well as a 48.84%reduction in maximum combinational path latency.展开更多
Low-power designs are getting increased significance in numerous applications like high-performance computing and wireless communication due to the rise in power dissipation.Power dissipation of VLSI(very large scale ...Low-power designs are getting increased significance in numerous applications like high-performance computing and wireless communication due to the rise in power dissipation.Power dissipation of VLSI(very large scale integration)circuits in test mode is much higher than in the normal operation mode due to the high frequency of applied test patterns.Product lifetime,yield,reduced performance,and circuit damage will result from this additional power consumption in testing.Therefore,the main objective of today’s test applications is to minimize power dissipation by increasing the correlation of applied successive test vectors.Low-power test pattern generator(TPG)using LFSR(linear feedback shift register)and binary to the excess-4 converter and binary ripple counter is proposed.The test vectors generated by the TPG have a high correlation between successive test vectors,which leads to minimum switching.During the testing of benchmark circuits,the proposed method shows a significant reduction in dynamic power consumption concerning its peer works.展开更多
The upgradation in Deep learning has enabled us to use the advantage of classification techniques into hyperspectral image for analysis of variety of image bands.In recent works,convolutional neural network(CNN)are pr...The upgradation in Deep learning has enabled us to use the advantage of classification techniques into hyperspectral image for analysis of variety of image bands.In recent works,convolutional neural network(CNN)are predominantly used for visual processing of data and its classification.Hyperspectral image classification deals with both spatial and spectral information extracted from the image datasets.3-D CNN is sparsely used due to their complexity in computation.However,it has the capability to facilitate both spatial and spectral data from the spectrum of remotely sensed images.Usage of 2-D CNN includes only the spatial information and cannot be feasible for classification.The proposed model includes 3-D CNN stacked with 2-D CNN reducing the complexity.This hybrid model validated for the remote sensing datasets of Indian Pines(IP),Pavia University(PU),Salinas(SA)and Kennedy Space Canter(KSC).The conventional principal component analysis(PCA)is applied to remove the spectral redundancy over the HSI datasets.The model incorporates both 2-D and 3-D CNN,collectively called the Hybrid Spectral CNN.It comprises of three 3-D convolutional layers and one 2-D convolution layer with 3FC(fully connected)layers.The results of the Hybrid model are compared with the 2-D CNN and 3-D CNN models.Several statistical tests like F1-score,Recall,Precision etc.,are also computed.All these experiments show that the performance of hyperspectral image classification is improved efficiently with this framework.展开更多
文摘The reinforced two layered foundation bed considered for study consists of a layer of granular fill overlying soft non-homogeneous clay with inclusion or reinforcement （geosymhetic strips, grids or sheets） in single layer at soil-granular fill interface A method is developed to estimate the bearing capacity of a strip footing on the surface of a reinforced foundation bed over a finite layer of clay whose undrained strength increases linearly with depth incorporating the contribution of axial resistance of the reinforcement together with those of granular fill and soft ground. Parametric studies presented quantify the improvement in bearing capacity.
文摘It is common knowledge that the end user of stream flow data may necessarily not have any prior knowledge of the quality control measures applied in their generation, therefore, conclusions drawn most often times may not be effective as desired. Thus, this study is an attempt at providing an independent quality construct to boost the confidence in the use of stream flow data by developing regional flow duration curves for selected ungauged stations of the upper Niger River Basin, Nigeria. Toward this end, stream flow data for seven gauging stations cover some sub basins in the Basin were obtained;precisely, monthly stream flow data covering a range of eleven to fifty-three years period. The flow duration curves from the gauging stations were fitted with three probability distribution models;i.e., logarithmic, power and exponential regression models. For the regionalisation, parameterisation was carried out in terms of the drainage area alone to allow for simplicity of models. Results obtained showed that the exponential regression model, in terms of Coefficient of Determination (R<sup>2</sup>) had the best fit. Though the regionalised model was simple, measurable agreement was obtained during the calibration and validation phases. However, considering the length of data used and probable variability in the stream flow regime, it is not possible to objectively generalise on the quality of the results. Against this backdrop, it suffices to take into cognisance the need to use an ensemble of catchment characteristics in the development of the flow duration curves and the overall regional models;this is important considering the implications of anthropogenic activities and hydro-climatic variations.
文摘India is an agricultural country and a core source of income for the world population.The Indian economy is greatly depending on agriculture that is decrease day by day due to pandemic COVID-19.India is a major exporter of many crop foods.India,Thailand,and Vietnam are the major exports of rice if these stopped exports it reduces the economy up to 15%.A related circumstance is built up with diverse yields too like wheat,sunflower whose fare has been stationary by Kazakhstan,Serbia individually.In India,the end of April is the main source of income to farmers because they sell their rabi crops(wheat,mustard,maize,lentil,chilies,gram,tomatoes)in the market drastically decreases of CFPI may lead to the distress of Indian agricultural economy.The change over time in the price of options on wheat futures reveals increased price volatility in response to growing uncertainty about the COVID-19 impacts.
基金We deeply acknowledge Taif University for supporting this study through Taif University Researchers Supporting Project Number(TURSP-2020/313),Taif University,Taif,Saudi Arabia.
文摘Internet of Things(IoT)is a recent paradigm to improve human lifestyle.Nowadays,number devices are connected to the Internet drastically.Thus,the people can control and monitor the physical things in real-time without delay.The IoT plays a vital role in all kind of fields in our world such as agriculture,livestock,transport,and healthcare,grid system,connected home,elderly people carrying system,cypher physical system,retail,and intelligent systems.In IoT energy conservation is a challenging task,as the devices are made up of low-cost and low-power sensing devices and local processing.IoT networks have significant challenges in two areas:network lifespan and energy usage.Therefore,the clustering is a right choice to prolong the energy in the network.In LEACH clustering protocol,sometimes the same node acts as CH again and again probabilistically.To overcome these issues,this paper proposes the Energy-Aware Cluster-based Routing(EACRLEACH)protocol in WSN based IoT.The Cluster Head(CH)selection is a crucial task in clustering protocol inWSN based IoT.In EACR-LEACH,the CH is selected by using the routing metrics,Residual Energy(RER),Number of Neighbors(NoN),Distance between Sensor Node and Sink(Distance)and Number of Time Node Act as CH(NTNACH).An extensive simulation is conducted on MATLAB 2019a.The accomplishment of EACR-LEACH is compared to LEACH and SE-LEACH.The proposed EACR-LEACH protocol extends the network’s lifetime by 4%-8%and boosts throughput by 16%–24%.
基金This research is funded by Taif University,TURSP-2020/313.
文摘In the present scenario,Deep Learning(DL)is one of the most popular research algorithms to increase the accuracy of data analysis.Due to intra-class differences and inter-class variation,image classification is one of the most difficult jobs in image processing.Plant or spinach recognition or classification is one of the deep learning applications through its leaf.Spinach is more critical for human skin,bone,and hair,etc.It provides vitamins,iron,minerals,and protein.It is beneficial for diet and is readily available in people’s surroundings.Many researchers have proposed various machine learning and deep learning algorithms to classify plant images more accurately in recent years.This paper presents a novel Convolutional Neural Network(CNN)to recognize spinach more accurately.The proposed CNN architecture classifies the spinach category,namely Amaranth leaves,Black nightshade,Curry leaves,and Drumstick leaves.The dataset contains 400 images with four classes,and each type has 100 images.The images were captured from the agricultural land located at Thirumanur,Salem district,Tamil Nadu.The proposed CNN achieves 97.5%classification accuracy.In addition,the performance of the proposed CNN is compared with Support Vector Machine(SVM),Random Forest,Visual Geometry Group 16(VGG16),Visual Geometry Group 19(VGG19)and Residual Network 50(ResNet50).The proposed provides superior performance than other models,namely SVM,Random Forest,VGG16,VGG19 and ResNet50.
文摘Food security for the 7 billion people on earth requires minimizing crop damage by timely detectionofdiseases.Most deep learningmodels forautomated detectionof diseases in plants suffer fromthe fatal flaw that once tested on independent data,their performance drops significantly.This work investigates a potential solution to this problem by using segmented image data to train the convolutional neural network(CNN)models.As compared to the F-CNN model trained using full images,S-CNN model trained using segmented imagesmore than doubles in performance to 98.6%accuracy when tested on independent data previously unseen by the models even with 10 disease classes.Not only this,by using tomato plant and target spot disease type as an example,we show that the confidence of self-classification for S-CNN model improves significantly over F-CNN model.This research work brings applicability of automated methods closer to non-experts for timely detection of diseases.
文摘Unbalanced operating condition in a power system can cause partial overloading of the generators in the network,a condition where one or two of the three phases of the generator become overloaded even if the total 3-phase power output of the generator is within its specified limit.Partial overloading of generators beyond certain limits is undesirable and must be avoided.Distribution systems are often subjected to highly unbalanced operating conditions.Introduction of distributed generations(DGs),therefore,has rendered today’s distribution systems quite susceptible to this problem.Mitigation of this problem requires the issue to be addressed properly during analysis,operation and planning of such systems.Analysis,operation and planning of power networks under unbalanced operating condition require 3-phase load flow study.The existing methods of 3-phase load flow are not equipped to take into account any limit on the loadings of the individual phases of the generators.In the present work,a methodology based on NewtonRaphson(N-R)3-phase load flow with necessary modifications is proposed.The proposed methodology is able to determine the safe loading limits of the generators,and,can be adopted for operation and planning of power networks under unbalanced operating conditions to overcome the above difficulties.Test results on IEEE-37 bus feeder network are presented to demonstrate the effectiveness of the proposed method.
文摘In the field of digital image processing,denoising is one of the basic problems.The challenges faced in image denoising are detecting impulse noise and designing a suitable filter.In this paper,we propose a methodology to remove the random impulse noise on the color image using a novel switching median filter.By using this novel technique,the occurrence of color artifacts has been avoided after noise removal which depends on auto-tuning threshold detection and a vector-type median filter noise remover.In the proposed technique,the random valued impulse noises with uniform distribution have been dealt with switching median filter.L_(2)Norm is employed to calculate the distribution distance rather than L1 Norm which is used to identify the optimal threshold value for auto-tuning filter.The switching auto-tuning detector automatically tunes the noisy pixels based on distance information of pixels distribution.The Normalized Mean Square Error(NMSE)is found to decrease for L_(2)Norm when compared with L1 Norm.The Peak Signal to Noise Ratio(PSNR)value and True Positive Rate(TPR)value improved with L_(2)Norm signifying effective noise removal.The efficiency of the present method is verified by conducting experiments on digital images.
基金funded by the research general direction at Universidad Santiago de Cali,Colombia under call no 01-2022.
文摘The current milieu,encourages rapid growth of wireless communication,multimedia applications,robotics and graphics to have efficient utilization of resources with high throughput and low power digital signal processing(DSP)systems.In an aggregate DSP system ranging from audio/video signal processing to wireless sensor networks,floating point matrix multiplication is used in wide scale in most of the fundamental processing units.Hardware implementation of floating-point matrix multiplication demands a colossal number of arithmetic operations that alter speed and consuming more area and power.DSP systems essentially uses two techniques to reduce dynamic power consumption:-they are pipelining and parallel processing that needs high performance processing element with less area and low power in diverse scientific computing applications.However,number of adders and multipliers used in the design of floating-point unit also increases subsequently.The adders and multipliers are the most area,delay and power consuming data path elements in the processing unit.The arithmetic level reduction of delay,power and area in the processing element is performed by the selection of appropriate adders and multipliers.This article proposes a parallel multiplication architecture using Strassen and UrdhvaTiryagbhyam multiplier,which involves design of efficient parallel matrix multiplication with flexible implementation of FPGA(Field Programmable Gate Array)device to analyse the computation and area.The design incorporates scheduling of blocks,operations on processing elements,block size determination,parallelization and double buffering for storage of matrix elements.
文摘The adaptive sign least mean square(SLMS)filter may change dynamically depending on the filter output.Noise cancellation is one of the most common adaptive filter applications.In real-time applications such as medical computing,the speed of the process producing hardware is critical,hence,this study proposes the hardware implementation of the SLMS adaptive filter utilizing the Xilinx System Generator.The suggested design seeks to improve performance while decreasing convergence rate and route latency.In this paper,(i)we proposed a modified architecture for an 8-tap SLMS adaptive filter is built,and(ii)multiplier-less structure for a Modified DLMS Filter is introduced and compare the same.The ECG signal was evaluated using the intended architecture.The algorithm’s functionality is tested in MATLAB using varied ECG data from the MIT-BIH database as input.Using Xilinx system generator,the LMS and SLMS are developed,simulated,synthesized,and implemented on a Virtex-5 FPGA.When comparing Systolic Sign LMS Filter to LMS Filter,the result reveals a 5%drop in total real-time router completion and a drop in the number of adders and subtractors,as well as a 48.84%reduction in maximum combinational path latency.
文摘Low-power designs are getting increased significance in numerous applications like high-performance computing and wireless communication due to the rise in power dissipation.Power dissipation of VLSI(very large scale integration)circuits in test mode is much higher than in the normal operation mode due to the high frequency of applied test patterns.Product lifetime,yield,reduced performance,and circuit damage will result from this additional power consumption in testing.Therefore,the main objective of today’s test applications is to minimize power dissipation by increasing the correlation of applied successive test vectors.Low-power test pattern generator(TPG)using LFSR(linear feedback shift register)and binary to the excess-4 converter and binary ripple counter is proposed.The test vectors generated by the TPG have a high correlation between successive test vectors,which leads to minimum switching.During the testing of benchmark circuits,the proposed method shows a significant reduction in dynamic power consumption concerning its peer works.
文摘The upgradation in Deep learning has enabled us to use the advantage of classification techniques into hyperspectral image for analysis of variety of image bands.In recent works,convolutional neural network(CNN)are predominantly used for visual processing of data and its classification.Hyperspectral image classification deals with both spatial and spectral information extracted from the image datasets.3-D CNN is sparsely used due to their complexity in computation.However,it has the capability to facilitate both spatial and spectral data from the spectrum of remotely sensed images.Usage of 2-D CNN includes only the spatial information and cannot be feasible for classification.The proposed model includes 3-D CNN stacked with 2-D CNN reducing the complexity.This hybrid model validated for the remote sensing datasets of Indian Pines(IP),Pavia University(PU),Salinas(SA)and Kennedy Space Canter(KSC).The conventional principal component analysis(PCA)is applied to remove the spectral redundancy over the HSI datasets.The model incorporates both 2-D and 3-D CNN,collectively called the Hybrid Spectral CNN.It comprises of three 3-D convolutional layers and one 2-D convolution layer with 3FC(fully connected)layers.The results of the Hybrid model are compared with the 2-D CNN and 3-D CNN models.Several statistical tests like F1-score,Recall,Precision etc.,are also computed.All these experiments show that the performance of hyperspectral image classification is improved efficiently with this framework.
