H.6.5.10. Remote sensing
Mehdi Alizadeh; Parvin Ahmadi; Masoumeh Azimzadeh
Abstract
Field boundary detection is a critical task in modern agriculture, enabling precision farming, optimized resource management, and efficient crop monitoring. Despite its importance, existing deep learning models often fail to achieve high accuracy in delineating field boundaries due to challenges such ...
Read More
Field boundary detection is a critical task in modern agriculture, enabling precision farming, optimized resource management, and efficient crop monitoring. Despite its importance, existing deep learning models often fail to achieve high accuracy in delineating field boundaries due to challenges such as complex landscapes, varying resolutions, and noise in remote sensing images. To overcome these limitations, we propose HURA-Net, an advanced deep learning framework that intelligently integrates UNet++, ResUNet, and an attention mechanism into a unified architecture. By hybridizing these models, HURA-Net effectively combines their strengths—such as multi-scale feature extraction (UNet++), residual learning (ResUNet), and focus on salient regions (attention mechanism)—while minimizing their individual weaknesses. To further enhance performance, we introduce a refined loss function that not only improves segmentation precision but also addresses the class imbalance problem, which is common in boundary detection tasks. Extensive experiments on a diverse dataset of high-resolution satellite images from different regions of Iran demonstrate that HURA-Net significantly outperforms existing state-of-the-art models. Specifically, it achieves a recall of 45.85% (a 15.59% improvement over ResUNet) and an F1-score of 42.62% (7.27% higher than ResUNet), setting a new benchmark for accuracy. Moreover, our study highlights the critical role of strategic data augmentation in boosting model generalization, particularly in handling variations in lighting, crop types, and field shapes. The success of HURA-Net underscores the importance of innovative architecture design, optimized loss functions, and robust training strategies in advancing remote sensing image segmentation.
D. Data
Zahra Ghorbani; Ali Ghorbanian
Abstract
Increasing the accuracy of time-series clustering while reducing execution time is a primary challenge in the field of time-series clustering. Researchers have recently applied approaches, such as the development of distance metrics and dimensionality reduction, to address this challenge. However, using ...
Read More
Increasing the accuracy of time-series clustering while reducing execution time is a primary challenge in the field of time-series clustering. Researchers have recently applied approaches, such as the development of distance metrics and dimensionality reduction, to address this challenge. However, using segmentation and ensemble clustering to solve this issue is a key aspect that has received less attention in previous research. In this study, an algorithm based on the selection and combination of the best segments created from a time-series dataset was developed. In the first step, the dataset was divided into segments of equal lengths. In the second step, each segment is clustered using a hierarchical clustering algorithm. In the third step, a genetic algorithm selects different segments and combines them using combinatorial clustering. The resulting clustering of the selected segments was selected as the final dataset clustering. At this stage, an internal clustering criterion evaluates and sorts the produced solutions. The proposed algorithm was executed on 82 different datasets in 10 repetitions. The results of the algorithm indicated an increase in the clustering efficiency of 3.07%, reaching a value of 67.40. The obtained results were evaluated based on the length of the time series and the type of dataset. In addition, the results were assessed using statistical tests with the six algorithms existing in the literature.
H.3.2.2. Computer vision
Masoumeh Esmaeiili; Kourosh Kiani
Abstract
The classification of emotions using electroencephalography (EEG) signals is inherently challenging due to the intricate nature of brain activity. Overcoming inconsistencies in EEG signals and establishing a universally applicable sentiment analysis model are essential objectives. This study introduces ...
Read More
The classification of emotions using electroencephalography (EEG) signals is inherently challenging due to the intricate nature of brain activity. Overcoming inconsistencies in EEG signals and establishing a universally applicable sentiment analysis model are essential objectives. This study introduces an innovative approach to cross-subject emotion recognition, employing a genetic algorithm (GA) to eliminate non-informative frames. Then, the optimal frames identified by the GA undergo spatial feature extraction using common spatial patterns (CSP) and the logarithm of variance. Subsequently, these features are input into a Transformer network to capture spatial-temporal features, and the emotion classification is executed using a fully connected (FC) layer with a Softmax activation function. Therefore, the innovations of this paper include using a limited number of channels for emotion classification without sacrificing accuracy, selecting optimal signal segments using the GA, and employing the Transformer network for high-accuracy and high-speed classification. The proposed method undergoes evaluation on two publicly accessible datasets, SEED and SEED-V, across two distinct scenarios. Notably, it attains mean accuracy rates of 99.96% and 99.51% in the cross-subject scenario, and 99.93% and 99.43% in the multi-subject scenario for the SEED and SEED-V datasets, respectively. Noteworthy is the outperformance of the proposed method over the state-of-the-art (SOTA) in both scenarios for both datasets, thus underscoring its superior efficacy. Additionally, comparing the accuracy of individual subjects with previous works in cross subject scenario further confirms the superiority of the proposed method for both datasets.
H.6. Pattern Recognition
A. Noruzi; M. Mahlouji; A. Shahidinejad
Abstract
A biometric system provides automatic identification of an individual based on a unique feature or characteristic possessed by him/her. Iris recognition (IR) is known to be the most reliable and accurate biometric identification system. The iris recognition system (IRS) consists of an automatic segmentation ...
Read More
A biometric system provides automatic identification of an individual based on a unique feature or characteristic possessed by him/her. Iris recognition (IR) is known to be the most reliable and accurate biometric identification system. The iris recognition system (IRS) consists of an automatic segmentation mechanism which is based on the Hough transform (HT). This paper presents a robust IRS in unconstrained environments. Through this method, first a photo is taken from the iris, then edge detection is done, later on a contrast adjustment is persecuted in pre-processing stage. Circular HT is subsequently utilized for localizing circular area of iris inner and outer boundaries. The purpose of this last stage is to find circles in imperfect image inputs. Also, through applying parabolic HT, boundaries are localized between upper and lower eyelids. The proposed method, in comparison with available IRSs, not only enjoys higher accuracy, but also competes with them in terms of processing time. Experimental results on images available in UBIRIS, CASIA and MMUI database show that the proposed method has an accuracy rate of 99.12%, 98.80% and 98.34%, respectively.
I. Computer Applications
M. Fateh; E. Kabir
Abstract
In this paper, we present a method for color reduction of Persian carpet cartoons that increases both speed and accuracy of editing. Carpet cartoons are in two categories: machine-printed and hand-drawn. Hand-drawn cartoons are divided into two groups: before and after discretization. The purpose of ...
Read More
In this paper, we present a method for color reduction of Persian carpet cartoons that increases both speed and accuracy of editing. Carpet cartoons are in two categories: machine-printed and hand-drawn. Hand-drawn cartoons are divided into two groups: before and after discretization. The purpose of this study is color reduction of hand-drawn cartoons before discretization. The proposed algorithm consists of the following steps: image segmentation, finding the color of each region, color reduction around the edges and final color reduction with C-means. The proposed method requires knowing the desired number of colors in any cartoon. In this method, the number of colors is not reduced to more than about 1.3 times of the desired number. Automatic color reduction is done in such a way that final manual editing to reach the desired colors is very easy.
