Mehdi Rezaeian

Mehdi Rezaeian

Auto-generated profile, based on publicly available information

Mehdi Rezaeian

Computer Engineering Department, Yazd University, Yazd, Iran

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2016 in Journal of Medical Signals & Sensors
Verified
Journal Article
Open access
PDF file
PMParinaz Mortaheb
Parinaz Mortaheb
MRMehdi Rezaeian
Mehdi Rezaeian
Segmentation and three-dimensional (3D) visualization of teeth in dental computerized tomography (CT) images are of dentists' requirements for both abnormalities diagnosis and the treatments such as dental implant and orthodontic planning. On the other hand, dental CT image segmentation is a difficult process because of the specific characteristics of the tooth's structure. This paper presents a method for automatic segmentation of dental CT images. We present a multi-step method, which starts with a preprocessing phase to reduce the metal artifact using the least square support vector machine. Integral intensity profile is then applied to detect each tooth's region candidates. Finally, the mean shift algorithm is used to partition the region of each tooth, and all these segmented slices are then applied for 3D visualization of teeth. Examining the performance of our proposed approach, a set of reliable assessment metrics is utilized. We applied the segmentation method on 14 cone-beam CT datasets. Functionality analysis of the proposed method demonstrated precise segmentation results on different sample slices. Accuracy analysis of the proposed method indicates that we can increase the sensitivity, specificity, precision, and accuracy of the segmentation results by 83.24%, 98.35%, 72.77%, and 97.62% and decrease the error rate by 2.34%. The experimental results show that the proposed approach performs well on different types of CT images and has better performance than all existing approaches. Moreover, segmentation results can be more accurate by using the proposed algorithm of metal artifact reduction in the preprocessing phase.

7 May 2020 in Current Medical Imaging
Verified
Journal Article
Closed access
PDF file
MKMahnaz Kazemipoor
Mahnaz Kazemipoor
MRMehdi Rezaeian
Mehdi Rezaeian
MKMaryam Kazemipoor
Maryam Kazemipoor
SHSareena Hamzah
Sareena Hamzah
SSShishir Kumar Shandilya
Shishir Kumar Shandilya
Background: Physical characteristics including body size and configuration, are considered as one of the key influences on the optimum performance in athletes. Despite several analyzing methods for modeling the slimming estimation in terms of reduction in anthropometric indices, there are still weaknesses of these models such as being very demanding including time taken for analysis and accuracy. Objective: This research proposes a novel approach for determining the slimming effect of a herbal composition as a natural medicine for weight loss. Methods: To build an effective prediction model, a modern hybrid approach, merging adaptivenetwork- based fuzzy inference system and particle swarm optimization (ANFIS-PSO) was constructed for prediction of changes in anthropometric indices including waist circumference, waist to hip ratio, thigh circumference and mid-upper arm circumference, on female athletes after consumption of caraway extract during ninety days clinical trial. Results: The outcomes showed that caraway extract intake was effective on lowering all anthropometric indices in female athletes after ninety days trial. The results of analysis by ANFIS-PSO was more accurate compared to SPSS. Also, the efficiency of the proposed approach was confirmed using the existing data. Conclusion: It is concluded that a development in predictive accuracy and simplification capability could be attained by hybrid adaptive neuro-fuzzy techniques as modern approaches in detecting changes in body characteristics. These developed techniques could be more useful and valid than other conventional analytical methods for clinical applications.

19 August 2021 in Multimedia Tools and Applications
Verified
Journal Article
Closed access
PDF file
SASoolmaz Abbasi
Soolmaz Abbasi
MRMehdi Rezaeian
Mehdi Rezaeian

1 January 2023 in Journal of Electronic Imaging
Verified
Journal Article
Closed access
PDF file
ZEZahra Esmaeily
Zahra Esmaeily
MRMehdi Rezaeian
Mehdi Rezaeian
Building vector extraction from aerial images is a challenge in many applications, especially location-based services. In recent years, different deep-learning techniques have improved the accuracy of building extraction. We propose a framework (JuliMa-Net) for building vector extraction from aerial images without boundary regularization or vertex sampling. The wireframe generated by our framework shows more detail about the building’s structure than its footprints. We initially selected three pretrained networks of Mask R-convolutional neural network, line, and junction detection. To improve the performance of the primary junction detection network, unique sets of decoders were designed. The smaller training set was used to fine-tune all three networks simultaneously. In addition, after applying proper processing on the obtained building masks, the outputs of the other two networks—the detected lines and junctions—were precisely selected and combined. The result of this process is a significant reduction in false detections and an increase in precision of 96%. Additionally, the final processing adds precise lines to the wireframes after combining junctions and lines. This improves the recall value.

28 April 2023 in JUCS - Journal of Universal Computer Science
Verified
Journal Article
Open access
PDF file
FRFahimeh Ramazankhani
Fahimeh Ramazankhani
MYMahdi Yazdian-Dehkord
Mahdi Yazdian-Dehkord
MRMehdi Rezaeian
Mehdi Rezaeian
Features extracted from facial images are used in various fields such as kinship verification. The kinship verification system determines the kin or non-kin relation between a pair of facial images by analysing their facial features. In this research, different texture and color features have been used along with the metric learning method, to verify the kinship for the four kinship relations of father-son, father-daughter, mother-son and mother-daughter. First, by fusing effective features, NRML metric learning used to generate the discriminative feature vector, then SVM classifier used to verify to kinship relations. To measure the accuracy of the proposed method, KinFaceW-I and KinFaceW-II databases have been used. The results of the evaluations show that the feature fusion and NRML metric learning methods have been able to improve the performance of the kinship verification system. In addition to the proposed approach, the effect of feature extraction from the image blocks or the whole image is investigated and the results are presented. The results indicate that feature extraction in block form, can be effective in improving the final accuracy of kinship verification.

8 June 2020 in IET Image Processing
Verified
Journal Article
Open access
PDF file
HAHamed Ayoobi
Hamed Ayoobi
MRMehdi Rezaeian
Mehdi Rezaeian
Loopy belief propagation (LBP) suffers from high computational time, specifically when each node in the Markov random field (MRF) model has lots of labels. In this study, a swift distance transformed belief propagation (SDT‐BP) method is proposed. SDT‐BP employs an efficient dynamic label pruning approach together with distance transformation to boost the running time of the LBP. The proposed dynamic label pruning approach is independent of any specific message scheduling. The resultant solution's energy is less than Priority‐BP. Furthermore, SDT‐BP guarantees convergence in fewer numbers of iterations. The direct combination of distance transformed belief propagation (DT‐BP) with the dynamic label pruning in Priority‐BP has O ( KTN log N ) computational complexity. However, the proposed method results in O ( KTN ) complexity. Where N is the number of nodes, K is the number of labels for each node, and T is the number of iterations. The authors conduct several experiments on image inpainting case studies, to evaluate this method. According to this analysis, DT‐BP faces nearly 90% speedup by preserving the energy of the solution at almost the same level. Furthermore, this method can be utilised in any MRF model where its distance function is transformable, i.e. in various image processing and computer vision problems.