In this article, we formalize the problem of semiautomatic construction of the contour of area objects from satellite hyperspectral images and present a solution algorithm using PCA and Dijkstra’s algorithm. The contour is considered as the boundary of an object, which can be used for its segmentation and classification. The semiautomatic contour accepts reference points specified by the operator. The formalization of the algorithm is completed.
The main scientific results of the 16th International Conference on Pattern Recognition and Information Processing (PRIP-2023), Minsk, Republic of Belarus, October 2023, are reviewed and analyzed. The history of this series of conferences is outlined, and its significant role in the development of the theory and practice of image analysis, pattern recognition, and artificial intelligence is indicated. A list of articles in the special issue is provided, prepared from reports selected by the PRIP-2023 Program Committee.
The article introduces a scientific gateway to assess land surface temperatures using Landsat 8 and visible infrared imaging radiometer suite data. The gateway offers a selection of four temperature retrieval algorithms and two interpolation methods to create time series. The evaluation of the gateway’s performance in Armenia from May to October 2022 is illustrated. The research identifies the Price, Jiménez-Muñoz, McMillin, and I05 Chanel algorithms as the most accurate nighttime temperature estimation. Additionally, these products exhibit a reasonable level of accuracy, with an average root mean squared error ranging from 2.42 to 2.45°C and a coefficient of determination spanning from 0.82 to 0.95. The outcomes of this study bear significant relevance for diverse applications such as urban heat island analysis, environmental monitoring, and agricultural assessments.
The paper proposes a new approach for crowd movement type estimation in video by combining convolutional neural network and integral optical flow. At first, main notions of crowd detection and tracking are given. Secondly, crowd movement features and parameters are defined. Three rules are proposed to identify direct crowd motion. Signs are presented for identifying chaotic crowd movement. Region movement indicators are introduced to analyze the movement of a group of people or a crowd. Thirdly, an algorithm of crowd movement types estimation using convolutional neural network and integral optical flow is proposed. We calculate crowd movement trajectories and show how they can be used to analyze behavior and divide crowds into groups of people. Experimental results show that with the help of convolutional neural network and integral optical flow crowd movement parameters can be calculated more accurately and quickly. The algorithm demonstrates stronger robustness to noise and the ability to get more accurate boundaries of moving objects.
The article presents an algebraic model for solving the problem of automation of ophthalmological diagnostics written in the language of descriptive image algebras. Descriptive image algebras are an initial mathematical language for formalizing and standardizing representations and procedures for processing image models and conversions over them when extracting information from images. To construct an algebraic model for solving the problem of automation of ophthalmological diagnostics, descriptive algebras of images with one ring are mainly used. This class of algebras belongs to the class of universal linear algebras with a sigma-associative ring with identity. A series of conversions and steps of the algebraic model are described using descriptive Boolean algebras over images. Descriptive image algebras are the main section of the mathematical apparatus of descriptive image analysis, which is a logically organized set of descriptive methods and models designed for image analysis and evaluation. The article defines specialized versions of descriptive image algebras with one ring and descriptive Boolean algebras over images, over models and representations of images, and over conversions of image models and images themselves, necessary for constructing an algebraic model. The image models (representations, formalized descriptions) used in writing the article are described. An example of a descriptive algorithmic scheme for solving an applied ophthalmological problem using an algebraic model is constructed.
The paper proposes the noninvasive image egg growing monitoring method based on an illumination and transfer learning. During the egg growing, the size of egg air cell is increased. The segmentation is performed to extract cells and segmentation parameters are adjusted and trained on an air cell datasets by transfer learning to separate air cells with high light transmittance from the background. The improved DeepLabV3+ network model for image egg monitoring is proposed. The network embeds coordinate attention in the lightweight network MobilenetV2. The decoder feature fusion method is improved to a semantic embedding branch structure. The middle-level features that have been newly introduced are merged with the high-level features and low-level features. The results show that the mean intersection over union of the model reaches 89.06% and that the mean pixel accuracy rate reaches 94.66%. The method can effectively segment the air cell part of the eggs. The feasibility of the method was verified by measuring the air cells of egg growing process from the 7th to the 19th day.
The paper deals with a state-of-art applied problem related to the neural networks training. It is shown that, given the expansion of the range of practical problems, gradient methods do not always satisfy the conditions of the subject area, which contributes to the development of alternative training methods. An original training algorithm is proposed that implements the annealing method, for which convergence to the optimal solution is proven. A modified version of the algorithm has been developed that is invariant to the size of the training sample. Experimental studies (using the example of solving problems of image classification and color image compression) confirm the effectiveness of the proposed approach.
A new solution to the problem of offline signature verification is presented. Digital images of signatures are processed and converted into a binary representation of a certain size. Then their contours are traced, and from them, two original features are calculated for describing the local structural features of the signature in the form of vectors of normalized frequency distributions of local binary pattern codes and values of local curvature of the signature contours. A new feature space is formed in which the pattern describes the proximity of pairs of signatures, and its coordinates are the rank correlation coefficients between the feature vectors of these signatures. In real practice, the expert has M (from 5 to 15) genuine signatures of a person; there are no forged signatures at all. On these M available genuine signatures of a single person, we train a one-class support vector machine model and obtain a single-writer-dependent classifier. A verifiable signature is considered forged if the classifier model considers it to be an outlier. The accuracy of our approach in verifying the genuineness of all 2640 signatures from the CEDAR database was 99.77%. All forged signatures in this database were correctly recognized.
The paper deals with a state-of-the-art applied problem related to the detection of landscape’s unnatural changes based on satellite images. An approach to constructing a detection system based on neural network processing of local terrain areas is proposed. As part of the approach, a neural network architecture and mechanisms for tuning to a specific area have been developed. It is shown that the use of neural networks and images corresponding to local areas (as initial data) provides easy expansion of the system to various types of terrain. The paper also presents a data filtering algorithm to adjust the balance of recall and overall precision of the system. Experimental studies have confirmed the effectiveness of the proposed approach.
In genome-wide association studies, combinations of single nucleotide polymorphisms are considered to be more effective than individual mutations in linking genes to traits. Clearly, finding the most relevant combinations from tens of thousands of these mutations associated with a trait is a complicated combinatorial problem. To achieve the higher prediction performance, improve computational efficiency and results interpretation, we proposed three algorithms for searching combinations of individual mutations and applied these algorithms to 3178 samples of Mycobacterium tuberculosis strains for predicting their drug resistance to 20 drugs. The single nucleotide polymorphisms associated with drug resistance were identified in the Mycobacterium tuberculosis genome using the single-marker test, and the combinations of individual mutations were searched using the multimarker test. The data were compared with those predicted by the widely recognized Mykrobe and TB-profiler software. Comparative analysis of the results obtained showed that, excepting for ofloxacin, the combinations of individual mutations found by our algorithms for the second-line drugs have some advantages in prediction accuracy.
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：