In Côte d’Ivoire, the recurring and unregulated use of bushfires, which cause ecological damage, presents a pressing concern for the custodians of protected areas. This study aims to enhance our comprehension of...In Côte d’Ivoire, the recurring and unregulated use of bushfires, which cause ecological damage, presents a pressing concern for the custodians of protected areas. This study aims to enhance our comprehension of the dynamics of burnt areas within the Abokouamékro Wildlife Reserve (AWR) by employing the analysis of spectral indices derived from satellite imagery. The research methodology began with the calculation of mean indices and their corresponding spectral sub-indices, including NDVI, SAVI, NDWI, NDMI, BAI, NBR, TCW, TCG, and TCB, utilizing data from the Sentinel-2A satellite image dated January 17, 2022. Subsequently, a fuzzy classification model was applied to these various indices and sub-indices, guided by the degree of membership α, with the goal of effectively distinguishing between burned and unburned areas. Following the classification, the accuracies of the classified indices and sub-indices were validated using the coordinates of 100 data points collected within the AWR through GPS technology. The results revealed that the overall accuracy of all indices and sub-indices declines as the degree of membership α decreases from 1 to 0. Among the mean spectral indices, NDVI-mean, SAVI-mean, NDMI-mean exhibited the highest overall accuracies, achieving 97%, 95%, and 90%, respectively. These results closely mirrored those obtained by sub-indices using band 8 (NDVI-B8, SAVI-B8, and NDMI-B8), which yield respective overall accuracies of 93%, 92%, and 89%. At a degree of membership α = 1, the estimated burned areas for the most effective indices encompassed 2144.38 hectares for NDVI-mean, 1932.14 hectares for mean SAVI-mean, and 4947.13 hectares for mean NDMI-mean. A prospective approach involving the amalgamation of these three indices could have the potential to yield improved outcomes. This study could be a substantial contribution to the discrimination of bushfires in Côte d’Ivoire.展开更多
Anemia is a blood abnormality that affects the quantity and quality of red blood cells in the human body. This sometimes banal sign spares no continent and no social stratum. This anomaly is generally appreciated thro...Anemia is a blood abnormality that affects the quantity and quality of red blood cells in the human body. This sometimes banal sign spares no continent and no social stratum. This anomaly is generally appreciated through biological analyzes of patients</span><span style="font-family:Verdana;">’</span><span style="font-family:Verdana;"> blood. These analyzes, which boil down to the knowledge of hemato-metric constants, cannot by themselves allow the characterization of certain forms of anemia in the sense that most anemia are related to the morphology and color of red blood cells. Our work in this paper is to perform blood smears on patients and perform a morphological and colorimetric analysis of red blood cells on these smears. This approach allowed us to highlight on each erythrocyte morphological and colorimetric descriptors to accurately identify the types of anemia by image processing methods. This identification is performed in an automated environment to allow pathologists to respond quickly to anemia-related emergencies and also improve the treatment to be conducted. This automation required the implementation of a new approach to electronic instrumentation and the acquisition of microscopic blood smear images for the automatic and rapid diagnosis of anemia.展开更多
In this work, we propose an original approach to the thin-layer identification of secondary metabolites (terpenes) based on the acquisition of multicomponent images integrating terpenes to be identified. Its principle...In this work, we propose an original approach to the thin-layer identification of secondary metabolites (terpenes) based on the acquisition of multicomponent images integrating terpenes to be identified. Its principle consists initially of segmentation by region of each component of the image based on the attribute tuples or colors of each region of the digital image. Then we proceeded to the calculations of region parameters such as standard deviation, entropy, average pixel color, eccentricity from an algorithm on the matlab software. These values allowed us to build a database. Finally, we built an algorithm for identifying secondary metabolites (terpenes) on the basis of these data. The relevance of our method of identifying or recognizing terpenes has been demonstrated compared to other methods, such as the one based on the calculation of frontal ratios which cannot discriminate between two terpenes having the same frontal ratio. The robustness of our method with respect to the identification of linalool, limonene was tested.展开更多
Image segmentation denotes a process for partitioning an image into distinct regions, it plays an important role in interpretation and decision making. A large variety of segmentation methods has been developed;among ...Image segmentation denotes a process for partitioning an image into distinct regions, it plays an important role in interpretation and decision making. A large variety of segmentation methods has been developed;among them, multidimensional histogram methods have been investigated but their implementation stays difficult due to the big size of histograms. We present an original method for segmenting n-D (where n is the number of components in image) images or multidimensional images in an unsupervised way using a fuzzy neighbourhood model. It is based on the hierarchical analysis of full n-D compact histograms integrating a fuzzy connected components labelling algorithm that we have realized in this work. Each peak of the histo- gram constitutes a class kernel, as soon as it encloses a number of pixels greater than or equal to a secondary arbitrary threshold knowing that a first threshold was set to define the degree of binary fuzzy similarity be- tween pixels. The use of a lossless compact n-D histogram allows a drastic reduction of the memory space necessary for coding it. As a consequence, the segmentation can be achieved without reducing the colors population of images in the classification step. It is shown that using n-D compact histograms, instead of 1-D and 2-D ones, leads to better segmentation results. Various images were segmented;the evaluation of the quality of segmentation in supervised and unsupervised of segmentation method proposed compare to the classification method k-means gives better results. It thus highlights the relevance of our approach, which can be used for solving many problems of segmentation.展开更多
In this work, we propose an original approach of semi-vectorial hybrid morphological segmentation for multicomponent images or multidimensional data by analyzing compact multidimensional histograms based on different ...In this work, we propose an original approach of semi-vectorial hybrid morphological segmentation for multicomponent images or multidimensional data by analyzing compact multidimensional histograms based on different orders. Its principle consists first of segment marginally each component of the multicomponent image into different numbers of classes fixed at K. The segmentation of each component of the image uses a scalar segmentation strategy by histogram analysis;we mainly count the methods by searching for peaks or modes of the histogram and those based on a multi-thresholding of the histogram. It is the latter that we have used in this paper, it relies particularly on the multi-thresholding method of OTSU. Then, in the case where i) each component of the image admits exactly K classes, K vector thresholds are constructed by an optimal pairing of which each component of the vector thresholds are those resulting from the marginal segmentations. In addition, the multidimensional compact histogram of the multicomponent image is computed and the attribute tuples or ‘colors’ of the histogram are ordered relative to the threshold vectors to produce (K + 1) intervals in the partial order giving rise to a segmentation of the multidimensional histogram into K classes. The remaining colors of the histogram are assigned to the closest class relative to their center of gravity. ii) In the contrary case, a vectorial spatial matching between the classes of the scalar components of the image is produced to obtain an over-segmentation, then an interclass fusion is performed to obtain a maximum of K classes. Indeed, the relevance of our segmentation method has been highlighted in relation to other methods, such as K-means, using unsupervised and supervised quantitative segmentation evaluation criteria. So the robustness of our method relatively to noise has been tested.展开更多
The mobile communication is nowadays one of the basic needs of humanity. It is essential to the flourishing of human beings. Considering this reality, the need to use its mobile phone is become more important and dive...The mobile communication is nowadays one of the basic needs of humanity. It is essential to the flourishing of human beings. Considering this reality, the need to use its mobile phone is become more important and diversified. The subscribers of the various mobile telephone operators are increasingly demanding. This situation poses the problems of the cover mobile network to the operators and leads them to opt for several solutions and investments. The mobile operators in order to satisfy their customers use a policy of pushing the limits of network coverage in time and space for festive moments in targeted zones. Thus, we have conducted a study on the topic: study of a kit of GSM radio site for event-driven movable coverage. This work is applied to GSM (Global system mobile) network of the operator Orange-Ci, leader of mobile telephony in Ivory Coast. We thus proceeded under investigation initially of the various aspects of the ordinary sites (motionless radio site) which are already deployed with Orange-Ci in order to impregnate us infrastructures and equipment used. This study revealed us that a radio site comprises 4 parts: infrastructures, installations and energy equipments, installations and radio equipments, and installations and equipment of transmission. After the first analysis, we made a study of the movable site. The study of the movable site enabled us to see the various possible solutions to fulfill the basic functions of a movable radio site. After analysis we retained that our radio site will be built on a truck on which a mast of 25 m maximum length for the antennas will be embarked, it will be fed by a generator also embarked on the truck and the solution of transmission selected is the transmission by satellite more precisely technology VSAT. We choose the various equipments (radio, transmission, energy) according to features which we defined to constitute the kit of movable radio site.展开更多
In this paper, we are presenting a new vector order, a solution to the open problem of the generalization of mathematical morphology to multicomponent images and multidimensional data. This approach uses the paradigm ...In this paper, we are presenting a new vector order, a solution to the open problem of the generalization of mathematical morphology to multicomponent images and multidimensional data. This approach uses the paradigm of P–order. Its primary principle consists, first in partitioning the multi-component image in the attribute space by a classification method in different numbers of classes, and then the vector attributes are ordered within each class (intra-order-class). And finally the classes themselves are ordered in turn from their barycenter (inter-class order). Thus, two attribute vectors (or colors) whatever, belonging to the vector image can be compared. Provided with this relation of order, vectors attributes of a multivariate image define a complete lattice ingredient necessary for the definition of the various morphological operators. In fact, this method creates a strong close similarity between vectors in order to move towards an order of the same principle as defined in the set of real numbers. The more the number of classes increases, the more the colors of the same class are similar and therefore the absolute adaptive referent tends to be optimal. On the other hand, the more the class number decreases or equals two, the more our approach tends towards the hybrid order developed previously. The proposed order has been implemented on different morphological operators through different multicomponent images. The fundamental robustness of our approach and that relating to noise have been tested. The results on the gradient, Laplacian and Median filter operators show the performance of our new order.展开更多
In this paper, it is proposed to apply the Dempster-Shafer Theory (DST) or the theory of evidence to map vegetation, aquatic and mineral surfaces with a view to detecting potential areas of observation of outcrops of ...In this paper, it is proposed to apply the Dempster-Shafer Theory (DST) or the theory of evidence to map vegetation, aquatic and mineral surfaces with a view to detecting potential areas of observation of outcrops of geological formations (rocks, breastplates, regolith, etc.). The proposed approach consists in aggregating information by using the DST. From pretreated Aster satellite images (geo-referencing, geometric correction and resampling at 15 m), new channels were produced by determining the spectral indices NDVI, MNDWI and NDBaI. Then, the DST formalism was modeled and generated under the MATLAB software, an image segmented into six classes including three absolute classes (E,V,M) and three classes of confusion ({E,V}, {M,V}, {E,M}). The control on the land, based on geographic coordinates of pixels of different classes on said image, has made it possible to make a concordant interpretation thereof. Our contribution lies in taking into account imperfections (inaccuracies and uncertainties) related to source information by using mass functions based on a simple support model (two focal elements: the discernment framework and the potential set of belonging of the pixel to be classified) with a normal law for the good management of these.展开更多
Image segmentation is an important stage in many applications such as image, video and computer processing. Generally image interpretation depends on it. The materials and methods used to demonstrate are described. Th...Image segmentation is an important stage in many applications such as image, video and computer processing. Generally image interpretation depends on it. The materials and methods used to demonstrate are described. The results are presented and analyzed. Several approaches and algorithms for image segmentation have been developed, but it is difficult to evaluate the efficiency and to make an objective comparison of different segmentation methods. This general problem has been addressed for the evaluation of a segmentation result and the results are available in the literature. In this work, we first presented some criteria of evaluation of segmentation commonly used in image processing with reviews of their models. Then multicomponent synthetic images of known composition are applied to these criteria to explore the operation and evaluate its relevance. The results show that choosing an assessment method depends on the purpose, however the criterion of Zeboudj appears powerful for the evaluation of region segmentations for properly separated classes, on the contrary the criteria of Levine-Nazif and Borsotti are adapted to the methods of classification and permit to build homogeneous regions or classes. The values of the Rosenbeger criterion are generally low and similar, so hard to make a comparison of segmentations with this criterion.展开更多
In this paper, the theory of plausible and paradoxical reasoning of Dezert- Smarandache (DSmT) is used to take into account the paradoxical charac-ter through the intersections of vegetation, aquatic and mineral surfa...In this paper, the theory of plausible and paradoxical reasoning of Dezert- Smarandache (DSmT) is used to take into account the paradoxical charac-ter through the intersections of vegetation, aquatic and mineral surfaces. In order to do this, we developed a classification model of pixels by aggregating information using the DSmT theory based on the PCR5 rule using the ∩NDVI, ∩MNDWI and ∩NDBaI spectral indices obtained from the ASTER satellite images. On the qualitative level, the model produced three simple classes for certain knowledge (E, V, M) and eight composite classes including two union classes characterizing partial ignorance ({E,V}, {M,V}) and six classes of intersection of which three classes of simple intersection (E∩V, M∩V, E∩M) and three classes of composite intersection (E∩{M,V}, M∩{E,V}, V∩{E,M}), which represent paradoxes. This model was validated with an average rate of 93.34% for the well-classified pixels and a compliance rate of the entities in the field of 96.37%. Thus, the model 1 retained provides 84.98% for the simple classes against 15.02% for the composite classes.展开更多
Deep learning has recently attracted a lot of attention with the aim of developing a fast, automatic and accurate system for image identification and classification. In this work, the focus was on transfer learning an...Deep learning has recently attracted a lot of attention with the aim of developing a fast, automatic and accurate system for image identification and classification. In this work, the focus was on transfer learning and evaluation of state-of-the-art VGG16 and 19 deep convolutional neural networks for fire image classification from fire images. In this study, five different approaches (Adagrad, Adam, AdaMax</span><span style="font-family:"">, </span><span style="font-family:"">Nadam and Rmsprop) based on the gradient descent methods used in parameter updating were studied. By selecting specific <span>learning rates, training image base proportions, number of recursion (epochs</span>), the advantages and disadvantages of each approach are compared with each <span>other in order to achieve the minimum cost function. The results of the comparison</span> are presented in the tables. In our experiment, Adam optimizers with the VGG16 architecture with 300 and 500 epochs tend to steadily improve their accuracy with increasing number of epochs without deteriorating performance. The optimizers were evaluated on the basis of their AUC of the ROC curve. It achieves a test accuracy of 96%, which puts it ahead of other architectures.展开更多
Any undesirable signal limiting to a degree or another the integrity and the intelligibility of a useful signal can be considered as noise. In the general rule, the good performance of a system is assured only if the ...Any undesirable signal limiting to a degree or another the integrity and the intelligibility of a useful signal can be considered as noise. In the general rule, the good performance of a system is assured only if the level of power of the useful signal exceeds by several orders of magnitude that of the noise (signal to noise of a several tens of decibels). However certain elaborate methods of treatment allow working with very low signal to noise ratio in an optimal way any a priori knowledge available on the signal useful to interpret. In this work, we evaluate the robustness of the noise on a new method of multicomponent image segmentation developed recently. Two types of additional noises are considered, which are the Gaussian noise and the uniform noise, with varying correlation between the different components (or planes) of the image. Quantitative results show the influence of the noise level on the segmentation method.展开更多
文摘In Côte d’Ivoire, the recurring and unregulated use of bushfires, which cause ecological damage, presents a pressing concern for the custodians of protected areas. This study aims to enhance our comprehension of the dynamics of burnt areas within the Abokouamékro Wildlife Reserve (AWR) by employing the analysis of spectral indices derived from satellite imagery. The research methodology began with the calculation of mean indices and their corresponding spectral sub-indices, including NDVI, SAVI, NDWI, NDMI, BAI, NBR, TCW, TCG, and TCB, utilizing data from the Sentinel-2A satellite image dated January 17, 2022. Subsequently, a fuzzy classification model was applied to these various indices and sub-indices, guided by the degree of membership α, with the goal of effectively distinguishing between burned and unburned areas. Following the classification, the accuracies of the classified indices and sub-indices were validated using the coordinates of 100 data points collected within the AWR through GPS technology. The results revealed that the overall accuracy of all indices and sub-indices declines as the degree of membership α decreases from 1 to 0. Among the mean spectral indices, NDVI-mean, SAVI-mean, NDMI-mean exhibited the highest overall accuracies, achieving 97%, 95%, and 90%, respectively. These results closely mirrored those obtained by sub-indices using band 8 (NDVI-B8, SAVI-B8, and NDMI-B8), which yield respective overall accuracies of 93%, 92%, and 89%. At a degree of membership α = 1, the estimated burned areas for the most effective indices encompassed 2144.38 hectares for NDVI-mean, 1932.14 hectares for mean SAVI-mean, and 4947.13 hectares for mean NDMI-mean. A prospective approach involving the amalgamation of these three indices could have the potential to yield improved outcomes. This study could be a substantial contribution to the discrimination of bushfires in Côte d’Ivoire.
文摘Anemia is a blood abnormality that affects the quantity and quality of red blood cells in the human body. This sometimes banal sign spares no continent and no social stratum. This anomaly is generally appreciated through biological analyzes of patients</span><span style="font-family:Verdana;">’</span><span style="font-family:Verdana;"> blood. These analyzes, which boil down to the knowledge of hemato-metric constants, cannot by themselves allow the characterization of certain forms of anemia in the sense that most anemia are related to the morphology and color of red blood cells. Our work in this paper is to perform blood smears on patients and perform a morphological and colorimetric analysis of red blood cells on these smears. This approach allowed us to highlight on each erythrocyte morphological and colorimetric descriptors to accurately identify the types of anemia by image processing methods. This identification is performed in an automated environment to allow pathologists to respond quickly to anemia-related emergencies and also improve the treatment to be conducted. This automation required the implementation of a new approach to electronic instrumentation and the acquisition of microscopic blood smear images for the automatic and rapid diagnosis of anemia.
文摘In this work, we propose an original approach to the thin-layer identification of secondary metabolites (terpenes) based on the acquisition of multicomponent images integrating terpenes to be identified. Its principle consists initially of segmentation by region of each component of the image based on the attribute tuples or colors of each region of the digital image. Then we proceeded to the calculations of region parameters such as standard deviation, entropy, average pixel color, eccentricity from an algorithm on the matlab software. These values allowed us to build a database. Finally, we built an algorithm for identifying secondary metabolites (terpenes) on the basis of these data. The relevance of our method of identifying or recognizing terpenes has been demonstrated compared to other methods, such as the one based on the calculation of frontal ratios which cannot discriminate between two terpenes having the same frontal ratio. The robustness of our method with respect to the identification of linalool, limonene was tested.
文摘Image segmentation denotes a process for partitioning an image into distinct regions, it plays an important role in interpretation and decision making. A large variety of segmentation methods has been developed;among them, multidimensional histogram methods have been investigated but their implementation stays difficult due to the big size of histograms. We present an original method for segmenting n-D (where n is the number of components in image) images or multidimensional images in an unsupervised way using a fuzzy neighbourhood model. It is based on the hierarchical analysis of full n-D compact histograms integrating a fuzzy connected components labelling algorithm that we have realized in this work. Each peak of the histo- gram constitutes a class kernel, as soon as it encloses a number of pixels greater than or equal to a secondary arbitrary threshold knowing that a first threshold was set to define the degree of binary fuzzy similarity be- tween pixels. The use of a lossless compact n-D histogram allows a drastic reduction of the memory space necessary for coding it. As a consequence, the segmentation can be achieved without reducing the colors population of images in the classification step. It is shown that using n-D compact histograms, instead of 1-D and 2-D ones, leads to better segmentation results. Various images were segmented;the evaluation of the quality of segmentation in supervised and unsupervised of segmentation method proposed compare to the classification method k-means gives better results. It thus highlights the relevance of our approach, which can be used for solving many problems of segmentation.
文摘In this work, we propose an original approach of semi-vectorial hybrid morphological segmentation for multicomponent images or multidimensional data by analyzing compact multidimensional histograms based on different orders. Its principle consists first of segment marginally each component of the multicomponent image into different numbers of classes fixed at K. The segmentation of each component of the image uses a scalar segmentation strategy by histogram analysis;we mainly count the methods by searching for peaks or modes of the histogram and those based on a multi-thresholding of the histogram. It is the latter that we have used in this paper, it relies particularly on the multi-thresholding method of OTSU. Then, in the case where i) each component of the image admits exactly K classes, K vector thresholds are constructed by an optimal pairing of which each component of the vector thresholds are those resulting from the marginal segmentations. In addition, the multidimensional compact histogram of the multicomponent image is computed and the attribute tuples or ‘colors’ of the histogram are ordered relative to the threshold vectors to produce (K + 1) intervals in the partial order giving rise to a segmentation of the multidimensional histogram into K classes. The remaining colors of the histogram are assigned to the closest class relative to their center of gravity. ii) In the contrary case, a vectorial spatial matching between the classes of the scalar components of the image is produced to obtain an over-segmentation, then an interclass fusion is performed to obtain a maximum of K classes. Indeed, the relevance of our segmentation method has been highlighted in relation to other methods, such as K-means, using unsupervised and supervised quantitative segmentation evaluation criteria. So the robustness of our method relatively to noise has been tested.
文摘The mobile communication is nowadays one of the basic needs of humanity. It is essential to the flourishing of human beings. Considering this reality, the need to use its mobile phone is become more important and diversified. The subscribers of the various mobile telephone operators are increasingly demanding. This situation poses the problems of the cover mobile network to the operators and leads them to opt for several solutions and investments. The mobile operators in order to satisfy their customers use a policy of pushing the limits of network coverage in time and space for festive moments in targeted zones. Thus, we have conducted a study on the topic: study of a kit of GSM radio site for event-driven movable coverage. This work is applied to GSM (Global system mobile) network of the operator Orange-Ci, leader of mobile telephony in Ivory Coast. We thus proceeded under investigation initially of the various aspects of the ordinary sites (motionless radio site) which are already deployed with Orange-Ci in order to impregnate us infrastructures and equipment used. This study revealed us that a radio site comprises 4 parts: infrastructures, installations and energy equipments, installations and radio equipments, and installations and equipment of transmission. After the first analysis, we made a study of the movable site. The study of the movable site enabled us to see the various possible solutions to fulfill the basic functions of a movable radio site. After analysis we retained that our radio site will be built on a truck on which a mast of 25 m maximum length for the antennas will be embarked, it will be fed by a generator also embarked on the truck and the solution of transmission selected is the transmission by satellite more precisely technology VSAT. We choose the various equipments (radio, transmission, energy) according to features which we defined to constitute the kit of movable radio site.
文摘In this paper, we are presenting a new vector order, a solution to the open problem of the generalization of mathematical morphology to multicomponent images and multidimensional data. This approach uses the paradigm of P–order. Its primary principle consists, first in partitioning the multi-component image in the attribute space by a classification method in different numbers of classes, and then the vector attributes are ordered within each class (intra-order-class). And finally the classes themselves are ordered in turn from their barycenter (inter-class order). Thus, two attribute vectors (or colors) whatever, belonging to the vector image can be compared. Provided with this relation of order, vectors attributes of a multivariate image define a complete lattice ingredient necessary for the definition of the various morphological operators. In fact, this method creates a strong close similarity between vectors in order to move towards an order of the same principle as defined in the set of real numbers. The more the number of classes increases, the more the colors of the same class are similar and therefore the absolute adaptive referent tends to be optimal. On the other hand, the more the class number decreases or equals two, the more our approach tends towards the hybrid order developed previously. The proposed order has been implemented on different morphological operators through different multicomponent images. The fundamental robustness of our approach and that relating to noise have been tested. The results on the gradient, Laplacian and Median filter operators show the performance of our new order.
文摘In this paper, it is proposed to apply the Dempster-Shafer Theory (DST) or the theory of evidence to map vegetation, aquatic and mineral surfaces with a view to detecting potential areas of observation of outcrops of geological formations (rocks, breastplates, regolith, etc.). The proposed approach consists in aggregating information by using the DST. From pretreated Aster satellite images (geo-referencing, geometric correction and resampling at 15 m), new channels were produced by determining the spectral indices NDVI, MNDWI and NDBaI. Then, the DST formalism was modeled and generated under the MATLAB software, an image segmented into six classes including three absolute classes (E,V,M) and three classes of confusion ({E,V}, {M,V}, {E,M}). The control on the land, based on geographic coordinates of pixels of different classes on said image, has made it possible to make a concordant interpretation thereof. Our contribution lies in taking into account imperfections (inaccuracies and uncertainties) related to source information by using mass functions based on a simple support model (two focal elements: the discernment framework and the potential set of belonging of the pixel to be classified) with a normal law for the good management of these.
文摘Image segmentation is an important stage in many applications such as image, video and computer processing. Generally image interpretation depends on it. The materials and methods used to demonstrate are described. The results are presented and analyzed. Several approaches and algorithms for image segmentation have been developed, but it is difficult to evaluate the efficiency and to make an objective comparison of different segmentation methods. This general problem has been addressed for the evaluation of a segmentation result and the results are available in the literature. In this work, we first presented some criteria of evaluation of segmentation commonly used in image processing with reviews of their models. Then multicomponent synthetic images of known composition are applied to these criteria to explore the operation and evaluate its relevance. The results show that choosing an assessment method depends on the purpose, however the criterion of Zeboudj appears powerful for the evaluation of region segmentations for properly separated classes, on the contrary the criteria of Levine-Nazif and Borsotti are adapted to the methods of classification and permit to build homogeneous regions or classes. The values of the Rosenbeger criterion are generally low and similar, so hard to make a comparison of segmentations with this criterion.
文摘In this paper, the theory of plausible and paradoxical reasoning of Dezert- Smarandache (DSmT) is used to take into account the paradoxical charac-ter through the intersections of vegetation, aquatic and mineral surfaces. In order to do this, we developed a classification model of pixels by aggregating information using the DSmT theory based on the PCR5 rule using the ∩NDVI, ∩MNDWI and ∩NDBaI spectral indices obtained from the ASTER satellite images. On the qualitative level, the model produced three simple classes for certain knowledge (E, V, M) and eight composite classes including two union classes characterizing partial ignorance ({E,V}, {M,V}) and six classes of intersection of which three classes of simple intersection (E∩V, M∩V, E∩M) and three classes of composite intersection (E∩{M,V}, M∩{E,V}, V∩{E,M}), which represent paradoxes. This model was validated with an average rate of 93.34% for the well-classified pixels and a compliance rate of the entities in the field of 96.37%. Thus, the model 1 retained provides 84.98% for the simple classes against 15.02% for the composite classes.
文摘Deep learning has recently attracted a lot of attention with the aim of developing a fast, automatic and accurate system for image identification and classification. In this work, the focus was on transfer learning and evaluation of state-of-the-art VGG16 and 19 deep convolutional neural networks for fire image classification from fire images. In this study, five different approaches (Adagrad, Adam, AdaMax</span><span style="font-family:"">, </span><span style="font-family:"">Nadam and Rmsprop) based on the gradient descent methods used in parameter updating were studied. By selecting specific <span>learning rates, training image base proportions, number of recursion (epochs</span>), the advantages and disadvantages of each approach are compared with each <span>other in order to achieve the minimum cost function. The results of the comparison</span> are presented in the tables. In our experiment, Adam optimizers with the VGG16 architecture with 300 and 500 epochs tend to steadily improve their accuracy with increasing number of epochs without deteriorating performance. The optimizers were evaluated on the basis of their AUC of the ROC curve. It achieves a test accuracy of 96%, which puts it ahead of other architectures.
文摘Any undesirable signal limiting to a degree or another the integrity and the intelligibility of a useful signal can be considered as noise. In the general rule, the good performance of a system is assured only if the level of power of the useful signal exceeds by several orders of magnitude that of the noise (signal to noise of a several tens of decibels). However certain elaborate methods of treatment allow working with very low signal to noise ratio in an optimal way any a priori knowledge available on the signal useful to interpret. In this work, we evaluate the robustness of the noise on a new method of multicomponent image segmentation developed recently. Two types of additional noises are considered, which are the Gaussian noise and the uniform noise, with varying correlation between the different components (or planes) of the image. Quantitative results show the influence of the noise level on the segmentation method.
