Ahmad Droby, Berat Kurar Barakat, Reem Alaasam, Boraq Madi, Irina Rabaev, Jihad El-Sana
Text line extraction is an essential preprocessing step in many handwritten document image analysis tasks. It includes detecting text lines in a document image and segmenting the regions of each detected line. Deep learning-based methods are frequently used for text line detection. However, only a limited number of methods tackle the problems of detection and segmentation together. This paper proposes a holistic method that applies Mask R-CNN for text line extraction. A Mask R-CNN model is trained to extract text lines fractions from document patches, which are further merged to form the text lines of an entire page. The presented method was evaluated on the two well-known datasets of historical documents, DIVA-HisDB and ICDAR 2015-HTR, and achieved state-of-the-art results. In addition, we introduce a new challenging dataset of Arabic historical manuscripts, VML-AHTE, where numerous diacritics are present. We show that the presented Mask R-CNN-based method can successfully segment text lines, even in such a challenging scenario.
{"title":"Text Line Extraction in Historical Documents Using Mask R-CNN","authors":"Ahmad Droby, Berat Kurar Barakat, Reem Alaasam, Boraq Madi, Irina Rabaev, Jihad El-Sana","doi":"10.3390/signals3030032","DOIUrl":"https://doi.org/10.3390/signals3030032","url":null,"abstract":"Text line extraction is an essential preprocessing step in many handwritten document image analysis tasks. It includes detecting text lines in a document image and segmenting the regions of each detected line. Deep learning-based methods are frequently used for text line detection. However, only a limited number of methods tackle the problems of detection and segmentation together. This paper proposes a holistic method that applies Mask R-CNN for text line extraction. A Mask R-CNN model is trained to extract text lines fractions from document patches, which are further merged to form the text lines of an entire page. The presented method was evaluated on the two well-known datasets of historical documents, DIVA-HisDB and ICDAR 2015-HTR, and achieved state-of-the-art results. In addition, we introduce a new challenging dataset of Arabic historical manuscripts, VML-AHTE, where numerous diacritics are present. We show that the presented Mask R-CNN-based method can successfully segment text lines, even in such a challenging scenario.","PeriodicalId":93815,"journal":{"name":"Signals","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44194137","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Plant classification requires the eye of an expert in botanics when the subtle differences in stem or petals differentiate between different species. Hence, an accurate automatic plant classification might be of great assistance to a person who studies agriculture, travels, or explores rare species. This paper focuses on a specific task of urban plants classification. The possible practical application of this work is a tool which assists people, growing plants at home, to recognize new species and to provide the relevant caring instructions. Because urban species are barely covered by the benchmark datasets, these species cannot be accurately recognized by the state-of-the-art pre-trained classification models. This paper introduces a new dataset, Urban Planter, for plant species classification with 1500 images categorized into 15 categories. The dataset contains 15 urban species, which can be grown at home in any climate (mostly desert) and are barely covered by existing datasets. We performed an extensive analysis of this dataset, aimed at answering the following research questions: (1) Does the Urban Planter dataset provide enough information to train accurate deep learning models? (2) Can pre-trained classification models be successfully applied on Urban Planter, and is the pre-training on ImageNet beneficial in comparison to the pre-training on a much smaller but more relevant dataset? (3) Does two-step transfer learning further improve the classification accuracy? We report the results of experiments designed to answer these questions. In addition, we provide the link to the installation code of the alpha version and the demo video of the web app for urban plants classification based on the best evaluated model. To conclude, our contribution is three-fold: (1) We introduce a new dataset of urban plant images; (2) We report the results of an extensive case study with several state-of-the-art deep networks and different configurations for transfer learning; (3) We provide a web application based on the best evaluated model. In addition, we believe that, by extending our dataset in the future to eatable plants and assisting people to grow food at home, our research contributes to achieve the United Nations’ 2030 Agenda for Sustainable Development.
{"title":"Urban Plants Classification Using Deep-Learning Methodology: A Case Study on a New Dataset","authors":"Marina Litvak, Sarit Divekar, Irina Rabaev","doi":"10.3390/signals3030031","DOIUrl":"https://doi.org/10.3390/signals3030031","url":null,"abstract":"Plant classification requires the eye of an expert in botanics when the subtle differences in stem or petals differentiate between different species. Hence, an accurate automatic plant classification might be of great assistance to a person who studies agriculture, travels, or explores rare species. This paper focuses on a specific task of urban plants classification. The possible practical application of this work is a tool which assists people, growing plants at home, to recognize new species and to provide the relevant caring instructions. Because urban species are barely covered by the benchmark datasets, these species cannot be accurately recognized by the state-of-the-art pre-trained classification models. This paper introduces a new dataset, Urban Planter, for plant species classification with 1500 images categorized into 15 categories. The dataset contains 15 urban species, which can be grown at home in any climate (mostly desert) and are barely covered by existing datasets. We performed an extensive analysis of this dataset, aimed at answering the following research questions: (1) Does the Urban Planter dataset provide enough information to train accurate deep learning models? (2) Can pre-trained classification models be successfully applied on Urban Planter, and is the pre-training on ImageNet beneficial in comparison to the pre-training on a much smaller but more relevant dataset? (3) Does two-step transfer learning further improve the classification accuracy? We report the results of experiments designed to answer these questions. In addition, we provide the link to the installation code of the alpha version and the demo video of the web app for urban plants classification based on the best evaluated model. To conclude, our contribution is three-fold: (1) We introduce a new dataset of urban plant images; (2) We report the results of an extensive case study with several state-of-the-art deep networks and different configurations for transfer learning; (3) We provide a web application based on the best evaluated model. In addition, we believe that, by extending our dataset in the future to eatable plants and assisting people to grow food at home, our research contributes to achieve the United Nations’ 2030 Agenda for Sustainable Development.","PeriodicalId":93815,"journal":{"name":"Signals","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47365358","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
George Voudiotis, Anna Moraiti, Sotirios Kontogiannis
One of the most critical causes of colony collapse disorder in beekeeping is caused by the Varroa mite. This paper presents an embedded camera module supported by a deep learning algorithm for the process of early detecting of Varroa infestations. This is achieved using a deep learning algorithm that tries to identify bees inside the brood frames carrying the mite in real-time. The end-node device camera module is placed inside the brood box. It is equipped with offline detection in remote areas of limited network coverage or online imagery data transmission and mite detection over the cloud. The proposed deep learning algorithm uses a deep learning network for bee object detection and an image processing step to identify the mite on the previously detected objects. Finally, the authors present their proof of concept experimentation of their approach that can offer a total bee and varroa detection accuracy of close to 70%. The authors present in detail and discuss their experimental results.
{"title":"Deep Learning Beehive Monitoring System for Early Detection of the Varroa Mite","authors":"George Voudiotis, Anna Moraiti, Sotirios Kontogiannis","doi":"10.3390/signals3030030","DOIUrl":"https://doi.org/10.3390/signals3030030","url":null,"abstract":"One of the most critical causes of colony collapse disorder in beekeeping is caused by the Varroa mite. This paper presents an embedded camera module supported by a deep learning algorithm for the process of early detecting of Varroa infestations. This is achieved using a deep learning algorithm that tries to identify bees inside the brood frames carrying the mite in real-time. The end-node device camera module is placed inside the brood box. It is equipped with offline detection in remote areas of limited network coverage or online imagery data transmission and mite detection over the cloud. The proposed deep learning algorithm uses a deep learning network for bee object detection and an image processing step to identify the mite on the previously detected objects. Finally, the authors present their proof of concept experimentation of their approach that can offer a total bee and varroa detection accuracy of close to 70%. The authors present in detail and discuss their experimental results.","PeriodicalId":93815,"journal":{"name":"Signals","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47745480","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yang Yang, Shihao Sun, Ao Chen, Siyang You, Yuqi Shen, Zhijun Li, Dayang Sun
The ranging error model is generally very complicated in actual ranging technologies. This paper gives an analysis of the biased distance substitution and proposes an unbiased multilateral positioning method to revise the biased substitution, making it an unbiased estimate of the squared distance. An unbiased estimate of the multilateral positioning formula is derived to solve the target node coordinates. Through simulation experiments, it is proved that the algorithm can improve the positioning accuracy, and the improvement is more obvious when the error variance is larger. Experiments using SX1280 also show that the ranging conforms to the biased error model, and the accuracy can be improved by using the unbiased estimator. When the actual experimental error standard deviation is 0.16 m, the accuracy can be improved by 0.15 m.
{"title":"The Analysis and Verification of Unbiased Estimator for Multilateral Positioning","authors":"Yang Yang, Shihao Sun, Ao Chen, Siyang You, Yuqi Shen, Zhijun Li, Dayang Sun","doi":"10.3390/signals3030029","DOIUrl":"https://doi.org/10.3390/signals3030029","url":null,"abstract":"The ranging error model is generally very complicated in actual ranging technologies. This paper gives an analysis of the biased distance substitution and proposes an unbiased multilateral positioning method to revise the biased substitution, making it an unbiased estimate of the squared distance. An unbiased estimate of the multilateral positioning formula is derived to solve the target node coordinates. Through simulation experiments, it is proved that the algorithm can improve the positioning accuracy, and the improvement is more obvious when the error variance is larger. Experiments using SX1280 also show that the ranging conforms to the biased error model, and the accuracy can be improved by using the unbiased estimator. When the actual experimental error standard deviation is 0.16 m, the accuracy can be improved by 0.15 m.","PeriodicalId":93815,"journal":{"name":"Signals","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46856617","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Research and development of image quality assessment (IQA) algorithms have been in the focus of the computer vision and image processing community for decades. The intent of IQA methods is to estimate the perceptual quality of digital images correlating as high as possible with human judgements. Full-reference image quality assessment algorithms, which have full access to the distortion-free images, usually contain two phases: local image quality estimation and pooling. Previous works have utilized visual saliency in the final pooling stage. In addition to this, visual saliency was utilized as weights in the weighted averaging of local image quality scores, emphasizing image regions that are salient to human observers. In contrast to this common practice, visual saliency is applied in the computation of local image quality in this study, based on the observation that local image quality is determined both by local image degradation and visual saliency simultaneously. Experimental results on KADID-10k, TID2013, TID2008, and CSIQ have shown that the proposed method was able to improve the state-of-the-art’s performance at low computational costs.
{"title":"Saliency-Guided Local Full-Reference Image Quality Assessment","authors":"D. Varga","doi":"10.3390/signals3030028","DOIUrl":"https://doi.org/10.3390/signals3030028","url":null,"abstract":"Research and development of image quality assessment (IQA) algorithms have been in the focus of the computer vision and image processing community for decades. The intent of IQA methods is to estimate the perceptual quality of digital images correlating as high as possible with human judgements. Full-reference image quality assessment algorithms, which have full access to the distortion-free images, usually contain two phases: local image quality estimation and pooling. Previous works have utilized visual saliency in the final pooling stage. In addition to this, visual saliency was utilized as weights in the weighted averaging of local image quality scores, emphasizing image regions that are salient to human observers. In contrast to this common practice, visual saliency is applied in the computation of local image quality in this study, based on the observation that local image quality is determined both by local image degradation and visual saliency simultaneously. Experimental results on KADID-10k, TID2013, TID2008, and CSIQ have shown that the proposed method was able to improve the state-of-the-art’s performance at low computational costs.","PeriodicalId":93815,"journal":{"name":"Signals","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45706970","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Transmission line fault classification forms the basis of fault protection management in power systems. Because faults have adverse effects on transmission lines, adequate measures must be implemented to avoid power outages. This paper focuses on using the categorical boosting (CatBoost) algorithm classifier to analyse and train multiple voltage and current data from a 330 kV and 500 km-long simulated faulty transmission line model designed using Matlab/Simulink. From it, 93,340 fault data sizes were extracted. The CatBoost classifier was employed to classify the faults after different machine learning algorithms were used to train the same data with different parameters. The trainer achieved the best accuracy of 99.54%, with an error of 0.46% for 748 iterations out of 1000. The algorithm was selected for its high performance in classifying faults based on accuracy, precision and speed. In addition, it is easy to use and handles multiple data-sets. In contrast, a support vector machine and an artificial neural network each has a longer training time than the proposed method’s 58.5 s. Proper fault classification techniques assist in the effective fault management and planning of power system control thereby preventing energy waste and providing high performance.
{"title":"Transmission Line Fault Classification of Multi-Dataset Using CatBoost Classifier","authors":"V. Ogar, Sajjad Hussain, K. Gamage","doi":"10.3390/signals3030027","DOIUrl":"https://doi.org/10.3390/signals3030027","url":null,"abstract":"Transmission line fault classification forms the basis of fault protection management in power systems. Because faults have adverse effects on transmission lines, adequate measures must be implemented to avoid power outages. This paper focuses on using the categorical boosting (CatBoost) algorithm classifier to analyse and train multiple voltage and current data from a 330 kV and 500 km-long simulated faulty transmission line model designed using Matlab/Simulink. From it, 93,340 fault data sizes were extracted. The CatBoost classifier was employed to classify the faults after different machine learning algorithms were used to train the same data with different parameters. The trainer achieved the best accuracy of 99.54%, with an error of 0.46% for 748 iterations out of 1000. The algorithm was selected for its high performance in classifying faults based on accuracy, precision and speed. In addition, it is easy to use and handles multiple data-sets. In contrast, a support vector machine and an artificial neural network each has a longer training time than the proposed method’s 58.5 s. Proper fault classification techniques assist in the effective fault management and planning of power system control thereby preventing energy waste and providing high performance.","PeriodicalId":93815,"journal":{"name":"Signals","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"69819160","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The nematode worm Caenorhabditis elegans has a relatively simple neural system for analysis of information transmission from sensory organ to muscle fiber. Consequently, this study includes an example of a neural circuit from the nematode worm, and a procedure is shown for measuring its information optimality by use of a logic gate model. This approach is useful where the assumptions are applicable for a neural circuit, and also for choosing between competing mathematical hypotheses that explain the function of a neural circuit. In this latter case, the logic gate model can estimate computational complexity and distinguish which of the mathematical models require fewer computations. In addition, the concept of information optimality is generalized to other biological systems, along with an extended discussion of its role in genetic-based pathways of organisms.
{"title":"A Perspective on Information Optimality in a Neural Circuit and Other Biological Systems","authors":"Robert Friedman","doi":"10.3390/signals3020025","DOIUrl":"https://doi.org/10.3390/signals3020025","url":null,"abstract":"The nematode worm Caenorhabditis elegans has a relatively simple neural system for analysis of information transmission from sensory organ to muscle fiber. Consequently, this study includes an example of a neural circuit from the nematode worm, and a procedure is shown for measuring its information optimality by use of a logic gate model. This approach is useful where the assumptions are applicable for a neural circuit, and also for choosing between competing mathematical hypotheses that explain the function of a neural circuit. In this latter case, the logic gate model can estimate computational complexity and distinguish which of the mathematical models require fewer computations. In addition, the concept of information optimality is generalized to other biological systems, along with an extended discussion of its role in genetic-based pathways of organisms.","PeriodicalId":93815,"journal":{"name":"Signals","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42129156","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Ana S. Santos Cardoso, R. L. Kæseler, M. Jochumsen, Lotte N. S. Andreasen Struijk
Brain–Computer Interfaces (BCIs) have been regarded as potential tools for individuals with severe motor disabilities, such as those with amyotrophic lateral sclerosis, that render interfaces that rely on movement unusable. This study aims to develop a dependent BCI system for manual end-point control of a robotic arm. A proof-of-concept system was devised using parieto-occipital alpha wave modulation and a cyclic menu with auditory cues. Users choose a movement to be executed and asynchronously stop said action when necessary. Tolerance intervals allowed users to cancel or confirm actions. Eight able-bodied subjects used the system to perform a pick-and-place task. To investigate the potential learning effects, the experiment was conducted twice over the course of two consecutive days. Subjects obtained satisfactory completion rates (84.0 ± 15.0% and 74.4 ± 34.5% for the first and second day, respectively) and high path efficiency (88.9 ± 11.7% and 92.2 ± 9.6%). Subjects took on average 439.7 ± 203.3 s to complete each task, but the robot was only in motion 10% of the time. There was no significant difference in performance between both days. The developed control scheme provided users with intuitive control, but a considerable amount of time is spent waiting for the right target (auditory cue). Implementing other brain signals may increase its speed.
{"title":"Manual 3D Control of an Assistive Robotic Manipulator Using Alpha Rhythms and an Auditory Menu: A Proof-of-Concept","authors":"Ana S. Santos Cardoso, R. L. Kæseler, M. Jochumsen, Lotte N. S. Andreasen Struijk","doi":"10.3390/signals3020024","DOIUrl":"https://doi.org/10.3390/signals3020024","url":null,"abstract":"Brain–Computer Interfaces (BCIs) have been regarded as potential tools for individuals with severe motor disabilities, such as those with amyotrophic lateral sclerosis, that render interfaces that rely on movement unusable. This study aims to develop a dependent BCI system for manual end-point control of a robotic arm. A proof-of-concept system was devised using parieto-occipital alpha wave modulation and a cyclic menu with auditory cues. Users choose a movement to be executed and asynchronously stop said action when necessary. Tolerance intervals allowed users to cancel or confirm actions. Eight able-bodied subjects used the system to perform a pick-and-place task. To investigate the potential learning effects, the experiment was conducted twice over the course of two consecutive days. Subjects obtained satisfactory completion rates (84.0 ± 15.0% and 74.4 ± 34.5% for the first and second day, respectively) and high path efficiency (88.9 ± 11.7% and 92.2 ± 9.6%). Subjects took on average 439.7 ± 203.3 s to complete each task, but the robot was only in motion 10% of the time. There was no significant difference in performance between both days. The developed control scheme provided users with intuitive control, but a considerable amount of time is spent waiting for the right target (auditory cue). Implementing other brain signals may increase its speed.","PeriodicalId":93815,"journal":{"name":"Signals","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48026774","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Houda Harkat, P. Monteiro, A. Gameiro, F. Guiomar, Hasmath Farhana Thariq Ahmed
MIMO-OFDM is a key technology and a strong candidate for 5G telecommunication systems. In the literature, there is no convenient survey study that rounds up all the necessary points to be investigated concerning such systems. The current deeper review paper inspects and interprets the state of the art and addresses several research axes related to MIMO-OFDM systems. Two topics have received special attention: MIMO waveforms and MIMO-OFDM channel estimation. The existing MIMO hardware and software innovations, in addition to the MIMO-OFDM equalization techniques, are discussed concisely. In the literature, only a few authors have discussed the MIMO channel estimation and modeling problems for a variety of MIMO systems. However, to the best of our knowledge, there has been until now no review paper specifically discussing the recent works concerning channel estimation and the equalization process for MIMO-OFDM systems. Hence, the current work focuses on analyzing the recently used algorithms in the field, which could be a rich reference for researchers. Moreover, some research perspectives are identified.
{"title":"A Survey on MIMO-OFDM Systems: Review of Recent Trends","authors":"Houda Harkat, P. Monteiro, A. Gameiro, F. Guiomar, Hasmath Farhana Thariq Ahmed","doi":"10.3390/signals3020023","DOIUrl":"https://doi.org/10.3390/signals3020023","url":null,"abstract":"MIMO-OFDM is a key technology and a strong candidate for 5G telecommunication systems. In the literature, there is no convenient survey study that rounds up all the necessary points to be investigated concerning such systems. The current deeper review paper inspects and interprets the state of the art and addresses several research axes related to MIMO-OFDM systems. Two topics have received special attention: MIMO waveforms and MIMO-OFDM channel estimation. The existing MIMO hardware and software innovations, in addition to the MIMO-OFDM equalization techniques, are discussed concisely. In the literature, only a few authors have discussed the MIMO channel estimation and modeling problems for a variety of MIMO systems. However, to the best of our knowledge, there has been until now no review paper specifically discussing the recent works concerning channel estimation and the equalization process for MIMO-OFDM systems. Hence, the current work focuses on analyzing the recently used algorithms in the field, which could be a rich reference for researchers. Moreover, some research perspectives are identified.","PeriodicalId":93815,"journal":{"name":"Signals","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48198172","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
L. Nanni, A. Lumini, Andrea Loreggia, A. Formaggio, Daniela Cuza
Recognizing objects in images requires complex skills that involve knowledge about the context and the ability to identify the borders of the objects. In computer vision, this task is called semantic segmentation and it pertains to the classification of each pixel in an image. The task is of main importance in many real-life scenarios: in autonomous vehicles, it allows the identification of objects surrounding the vehicle; in medical diagnosis, it improves the ability of early detecting of dangerous pathologies and thus mitigates the risk of serious consequences. In this work, we propose a new ensemble method able to solve the semantic segmentation task. The model is based on convolutional neural networks (CNNs) and transformers. An ensemble uses many different models whose predictions are aggregated to form the output of the ensemble system. The performance and quality of the ensemble prediction are strongly connected with some factors; one of the most important is the diversity among individual models. In our approach, this is enforced by adopting different loss functions and testing different data augmentations. We developed the proposed method by combining DeepLabV3+, HarDNet-MSEG, and Pyramid Vision Transformers. The developed solution was then assessed through an extensive empirical evaluation in five different scenarios: polyp detection, skin detection, leukocytes recognition, environmental microorganism detection, and butterfly recognition. The model provides state-of-the-art results.
{"title":"An Empirical Study on Ensemble of Segmentation Approaches","authors":"L. Nanni, A. Lumini, Andrea Loreggia, A. Formaggio, Daniela Cuza","doi":"10.3390/signals3020022","DOIUrl":"https://doi.org/10.3390/signals3020022","url":null,"abstract":"Recognizing objects in images requires complex skills that involve knowledge about the context and the ability to identify the borders of the objects. In computer vision, this task is called semantic segmentation and it pertains to the classification of each pixel in an image. The task is of main importance in many real-life scenarios: in autonomous vehicles, it allows the identification of objects surrounding the vehicle; in medical diagnosis, it improves the ability of early detecting of dangerous pathologies and thus mitigates the risk of serious consequences. In this work, we propose a new ensemble method able to solve the semantic segmentation task. The model is based on convolutional neural networks (CNNs) and transformers. An ensemble uses many different models whose predictions are aggregated to form the output of the ensemble system. The performance and quality of the ensemble prediction are strongly connected with some factors; one of the most important is the diversity among individual models. In our approach, this is enforced by adopting different loss functions and testing different data augmentations. We developed the proposed method by combining DeepLabV3+, HarDNet-MSEG, and Pyramid Vision Transformers. The developed solution was then assessed through an extensive empirical evaluation in five different scenarios: polyp detection, skin detection, leukocytes recognition, environmental microorganism detection, and butterfly recognition. The model provides state-of-the-art results.","PeriodicalId":93815,"journal":{"name":"Signals","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48476891","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：