FYFemin Yalcin

Femin Yalcin

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Femin Yalcin has a total of 34 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    A Topic Modeling Perspective for Emerging Trends Discovery in Hydrology: A Roadmap

    by Sila Ovgu Korkut, Oznur Oztunc-Kaymak, Aytug Onan, Erman Ulker, Femin Yalcin

    Determining trends can be crucial in guiding researchers, scientists, and agencies in the new knowledge age. The most challenging task is keeping track of the latest subjects among the stacks of publications. Thus, it is essential to do research that suggests automatic trending themes in an area to anticipate future subjects. This study aims to assess the emerging topics in the field of "hydrology" using topic modeling approach. The methodology of the current study comprises three key components: data collection, preprocessing and extracting significant features of research articles, and, finally, assessing trend topics in the field. To do so, latent semantic analysis (LSA), latent Dirichlet Allocation (LDA), and non-negative matrix factorization (NMF) have been considered. The comparative analysis of the coherence score shows that the NMF model outperformed the other methods. Based on the results of the NMF model, the topics in 2022 are "climate change", "hydrologic cycle", "natural hazards and erosion", "river basin", and "water management". The obtained topics were evaluated by an expert in the field. It is observed that the topics of 2022 kept their importance in 2023. The methodology presented in this study can be easily imitated in other fields. Moreover, based on the trend of the hydrology researchers’ studies, this study can provide perspectives for the government to take precautions.

  • Open Access

    DT-Inspect: A Deployment-Oriented, Training-Time-Only Image-Enhancement Framework forWind Turbine Drivetrain Damage Detection from Borescope Field Imagery

    by Oğuzhan Yavuz, Çağdaş Karataş, Femin Yalcin

    Wind turbine drivetrains operate under extreme mechanical excitations, and gearbox failures cause the longest downtime per failure event, frequently weeks of outage, and direct costs exceeding several hundred thousand dollars per incident. As much of the global fleet reaches its design life, exposure to undetected drivetrain damage is intensifying, making early and repeatable diagnosis a prerequisite. The standard diagnostic tool, borescopic visual inspection, faces demanding field conditions that degrade standard object detectors. Consequently, triage of inspection frames remains a manual and labor-intensive task. Existing computer-vision approaches either depend on inference-time restoration that increases latency, or rely on architecture-side redesigns that break the inference-cost envelope of borescope deployment. We therefore propose DT-Inspect, the first object-detection-based decision-support framework for drivetrain visual inspection combining training-time-only image enhancement, an ANSI/AGMA 1010 and ISO 10825-grounded real-world drivetrain borescope dataset, and a class-asymmetric loss intervention in a single deployment-realistic pipeline. The enhancement modules are active exclusively during training and impose zero computational overhead at inference, leaving the deployed model identical to vanilla YOLOv8s in parameter count, FLOPs and latency. Experiments against six two-stage R-CNN derivatives and eleven one-stage detectors demonstrate that DT-Inspect achieves validation Precision 0.953, Recall 0.802, mAP@0.50 0.871 and mAP@0.50:0.95 0.695, with mean IoU 0.826 and median IoU 0.914 on a held-out test set, while preserving the inference cost of the baseline detector. To close the gap between deep-learning research and industrial maintenance practice, the framework is operationalized through a lightweight desktop application that runs locally on an operator-laptop and emits AGMA/ISO-conformant annotated images.

  • Cumulative Residual Fisher Information Based on Finite and Infinite Mixture Models

    by Omid Kharazmi, Femin Yalcin, Javier E. Contreras-Reyes

    In this work, we extend the concept of Cumulative Residual Fisher (CRF) information through finite and infinite mixture survival functions. We first examine this information measure for infinite mixture survival functions and introduce the Jensen-cumulative residual Fisher (Jensen-CRF) information for these infinite mixture models. We show that this Jensen-type information measure can be expressed via mixture of an infinite number of relative cumulative residual information measures. We then propose the generalized cumulative residual Fisher (generalized CRF) information, emphasizing its convexity property, supported by a key lemma ensuring this trait. Next, we consider the q-cumulative residual Fisher (q-CRF) information and investigate its convexity property through a significant lemma. Further, we define the Jensen-q-cumulative residual Fisher (Jensen-q-CRF) divergence and relative q-cumulative residual Fisher (relative q-CRF) information in the case of a finite mixture model. Finally, we apply the Jensen-q-CRF divergence to image processing. Our numerical results related to image quality assessment show that the Jensen-q-CRF divergence measure effectively quantifies image similarity.


  • Machine learning algorithms for building height estimations using ICESat-2/ATLAS and Airborne LiDAR data

    by Muge Agca, Aslıhan Yucel, Efdal Kaya, Ali İhSan Daloglu, Mert Kayalık, Mevlut Yetkin, Femin Yalcın


  • Open Access

    Investigation and Comparison of Climate Boundary Maps Generated with Various Climate Classifications

    by Femin Yalçın, Deniz Arca, Kamil Kaygusuz, Uğur Büyük, Emre Demir, Sayit Özbey, İsmet Tıkız, Şeyma Nur Çevik, Sezin Güleryüz, Cemile Emel Yaz, et al.

    Climate has continuously undergone change throughout history. Understanding this change contributes significantly to the determination of boundaries for different climate types. Climate classifications aim to differentiate various climate types and identify similar or different locations. This facilitates the sustainable use of local resources and the development of land use plans. Additionally, comprehending the impact of climate on property is crucial for its management and regulation, enabling a more organized approach to property utilization. In this context, various climate classification methods have been developed. These methods assist in determining local climate differences, tracking changes over time, and establishing boundaries suitable for different climate types. This study aims to create climate boundary maps for Burdur province using Köppen, Trewartha, de Martonne, Aydeniz, Erinç, and Thornthwaite methods. Within the scope of the study, data from 11 meteorological observation stations in Burdur province have been organized within a Geographic Information System (GIS) and classified into climate types. Finally, climate boundary maps representing the entire region were generated using the Kriging interpolation method based on the identified climate classification.


  • Open Access

    Advanced Laser Material Processing Techniques

    by Sayit Özbey, İsMet Tıkız, Kamil Kaygusuz, Uğur Büyük, Femin Yalçın, Deniz Arca, Emre Demir, Şeyma Nur Çevik, Sezin Güleryüz, Cemile Emel Yaz, et al.

    The growing importance of laser material processing technologies in various industries, expanding application areas, and decreasing costs of laser systems make this technology critically important. This paper provides a comprehensive review of the advances, applications and impacts of laser technology in manufacturing, with a particular focus on laser surface treatment, welding, cutting, drilling and cladding. Academic research in this area is leading to the development of innovative manufacturing techniques aimed at improving product quality, designing multi-material components and supporting economic benefits. Numerous studies have been conducted to investigate and optimize the effects of lasers on materials, leading to significant advances in laser materials processing. Key findings highlight the importance of laser surface treatment in enhancing material properties, the versatility and precision provided by laser welding, the advantages of non-contact machining, the high speed and flexibility of laser cutting, and the capacity of laser drilling to effectively process hard and high-strength materials. Furthermore, it is critical to carefully determine the appropriate laser parameters to achieve the desired mechanical properties in laser-processed materials. Ongoing research is directed toward further understanding laser-material interactions and improving laser processing techniques. In brief, laser material processing technologies continue to play an important role in improving manufacturing processes and enhancing product quality.