The system of automatic license plate recognition (ALPR) is a combination of software and hardware technologies implementing ALPR algorithms. It seems to be easy to achieve the goal but recognition of license plate requires many difficult solutions to some non-trivial tasks. If the license plate is oriented horizontally, uniformly lighted, has a clean surface, clearly distinguishable characters, then it’ll be not too difficult to recognize such a license plate. However, the reality is much worse. The lighting of each part of the plate isn’t equal; the picture from the camera is noisy. Besides, the license plate can have a big angle relative to the camera and be dirty. These obstacles make it difficult to recognize the license plate characters and determine their location on the image. For instance, the accuracy of recognition is much worse on large camera angles. To solve these problems, the developers of automatic license plate recognition systems use a different approach to processing and analysis of images. The work shows an automatic license plate recognition system, which increases the recognition accuracy at large camera angles. The system is based on the technology of recognition of images with the use of highly accurate convolutional neural networks. The proposed system improves stages of normalization and segmentation of an image of the license plate, taking on large camera angles. The goal of improvements is to increase of accuracy of recognition. On the stage of normalization, before histogram equalization, the affine transformation of the image is performed. For the process of segmentation and recognition, Mask R-CNN is used. As the main segment-search algorithm, selective search is chosen. The combined loss function is used to fasten the process of training and classification of the network. The additional module to the convolutional neural network is added for solving the interclass segmentation. The input for this module is generated feature tensor. The output is segmented data for semantic processing. The developed system was compared to well-known systems (SeeAuto.USA and Nomeroff.Net). The invented system got better results on large camera shooting angles.
In recent times, machine learning algorithms have shown great performance in solving problems in different fields of study, including the analysis of remote sensing images, computer vision, natural language processing, medical issues, etc. A well-prepared input dataset can have a huge impact on the result metrics. However, a correctly selected hyperparameter combined with neural network architecture could highly increase the final metrics. Therefore, the hyperparameters optimization problem becomes a key issue in a deep learning algorithm. The process of finding a suitable hyperparameter combination could be performed manually or automatically. Manual search is based on previous research and requires enormous human efforts. However, there are many automated hyperparameter optimization methods have been successfully applied in practice. The automated hyperparameter tuning techniques are divided into two groups: black-box optimization techniques (such as Grid Search, Random Search) and multi-fidelity optimization techniques (HyperBand, BOHB). The most recent and promising among all approaches is BOHB which, which combines both Bayesian optimization and bandit-based methods, outperforms classical approaches, and can run asynchronously with given GPU resources and time budget that plays a vital role in the hyperparameter optimization process. The previous study proposed a convolutional deep learning neural network for solving land cover classification problems in the EuroSAT dataset. It was found that adding spectral indexes NDVI, NDWI, and GNDVI with RGB channels increased the result accuracy (from 64.72% to 84.19%) and F1 (from 63.89 % to 84.05%) score. However, the convolutional neural network architecture and hyperparameter combination were selected manually. The research optimizes convolutional neural network architecture and finds suitable hyperparameter combinations applied to land cover classification problems using multispectral images. The obtained results must increase result performance compared with the previous study and given budget constraints.
The subject matter of the article is the method of automating the identification of vehicle plate numbers based on the processing of one-aspect images obtained using video recording means. The goal is to provide automation of the process of identifying vehicle plate numbers within a wide range of changing the viewing angles and the levels of illumination. The task is formulation of the method of automated identification of vehicle plate numbers on one-aspect images, which are obtained by means of video fixation within wide limits of changing both the viewing angles and the levels of illumination. Analysis of the problems of methods and algorithms of automated detection and recognition of vehicle plate numbers has shown that it is most promising to use flexible algorithms that adapt to the changing conditions of observation of traffic control devices. One of the promising technologies for implementing such algorithms is the application of artificial neural networks. The solution of the problem of recognition of vehicle plate numbers can be represented as a complex of image processing and analysis of algorithms, which includes the initial preparation of the image, the discovery of the area of the vehicle plate on the image, the segmentation of symbols and the recognition of symbols. Conclusions: an algorithmically implemented method of identifying vehicle plate numbers, which makes possible searching the text areas under an arbitrary angle in different lighting conditions, is proposed. This method allows automating the process of identification of vehicle plate numbers within a wide range of distances to the car, as well as viewing the angles and levels of illumination. The purpose of further research is to improve the proposed method for its implementation, using modern software and hardware. K e ywor d s : vehicle plate number identification; image processing; character recognition; neural network; convolutional neural network.
Among all types of cancer, breast cancer is the most common. In 2017 breast cancer was the fourth rate for death reasons in Ukraine. The paper is devoted to the automatization of histopathological analysis, which can improve the process of cancer stage diagnosis. The purpose of the paper is to research the ability to use convolutional neural networks for classifying biopsy images for cancer diagnosis. The tasks of research are: analyzing cancer statistics in Europe and Ukraine; analyzing usage of Machine Learning in cancer prognosis and diagnosis tasks; preprocessing of BreCaHAD dataset images; developing a convolutional neural network and analyzing results; the building of heatmap. The object of the research is the process of detecting tumors in microscopic biopsy images using Convolutional Neural Network. The subject of the research is the process of classifying healthy and cancerous cells using deep learning neural networks. The scientific novelty of the research is using ConvNet trained on the BreCaHAD dataset for histopathological analysis. The theory of deep learning neural networks and mathematical statistics methods are used. In result it is obtained that the classification accuracy for a convolutional neural network on the test data is 0.935, ConvNet was effectively used for heatmap building. K e ywor d s : deep learning; convolutional neural networks; breast cancer; histopathological analysis; biopsy images; BreCaHAD.
The subject of study in the article is artificial intelligence methods that can be used for recognition of specific areas of the earth's surface in multispectral images provided by Earth remote sensing systems (ERS). The goal is to automate data analysis for recognizing areas affected by fire on multispectral remote sensing images. The task is to study and formulate a method for processing multispectral data, which makes it possible to automate the process of operational recognition of areas of burned-out areas in images, for the development of an eco-monitoring software system using artificial intelligence tools such as deep learning and neural networks. As a result of the analysis of modern methods of processing multispectral data, an investigation of the supervised learning strategy was chosen. The choice of the described method for solving an applied problem is based on the high flexibility of these method, as well as, provided that there is a sufficient amount of used training input data and correct training strategies, the possibility of analyzing heterogeneous multispectral data with ensuring high accuracy of results for each individual sample. Conclusions: the application of methodologies for intelligent processing of multispectral images has been investigated and substantiated. The theoretical foundations of the construction of neural networks are considered, the applied area of application is selected. An architectural model of a software product is analyzed and proposed, taking into account its scalability, the model of software system is developed and the results of its work are shown. The obtained results show the efficiency of proposed system and prospects of the proposed algorithms, which is a reason for further research and improvement of the used algorithms, with their possible use in industrial and enterprise eco-monitoring systems.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.