Search

This study focuses on the neural machine translation task for the TR-EN language pair, which is considered a low-resource language pair. We investigated fine-tuning strategies for pre-trained language models. Specifically, we explored the effectiveness of parameter-efficient adapter methods for fine-tuning multilingual pre-trained language models. Various combinations of LoRA and bottleneck adapters were experimented with. The combination of LoRA and bottleneck adapters demonstrated superior performance compared to other methods. This combination required only 5% of the pre-trained language model to be fine-tuned. The proposed method enhances parameter efficiency and reduces computational costs. Compared to the full fine-tuning of the multilingual pre-trained language model, it showed only a 3% difference in the BLEU score. Thus, nearly the same performance was achieved at a significantly lower cost. Additionally, models using only bottleneck adapters performed worse despite having a higher parameter count. Although adding LoRA to pre-trained language models alone did not yield sufficient performance, the proposed method improved machine translation. The results obtained are promising, particularly for low-resource language pairs. The proposed method requires less memory and computational load while maintaining translation quality. [ABSTRACT FROM AUTHOR]

As information systems become more widespread, data security becomes increasingly important. While traditional encryption methods provide effective protection against unauthorized access, they often struggle with multimedia data like images and videos. This necessitates specialized image encryption approaches. With the rise of mobile and Internet of Things (IoT) devices, lightweight image encryption algorithms are crucial for resource-constrained environments. These algorithms have applications in various domains, including medical imaging and surveillance systems. However, the biggest challenge of lightweight algorithms is balancing strong security with limited hardware resources. This work introduces a novel nonlinear matrix permutation approach applicable to both confusion and diffusion phases in lightweight image encryption. The proposed method utilizes three different chaotic maps in harmony, namely a 2D Zaslavsky map, 1D Chebyshev map, and 1D logistic map, to generate number sequences for permutation and diffusion. Evaluation using various metrics confirms the method's efficiency and its potential as a robust encryption framework. The proposed scheme was tested with 14 color images in the SIPI dataset. This approach achieves high performance by processing each image in just one iteration. The developed scheme offers a significant advantage over its alternatives, with an average NPCR of 99.6122, UACI of 33.4690, and information entropy of 7.9993 for 14 test images, with an average correlation value as low as 0.0006 and a vast key space of 2 800 . The evaluation results demonstrated that the proposed approach is a viable and effective alternative for lightweight image encryption. [ABSTRACT FROM AUTHOR]

Histopathology, vital in diagnosing medical conditions, especially in cancer research, relies on analyzing histopathology images (HIs). Nuclei segmentation, a key task, involves precisely identifying cell nuclei boundaries. Manual segmentation by pathologists is time‐consuming, prompting the need for robust automated methods. Challenges in segmentation arise from HI complexities, necessitating advanced techniques. Recent advancements in deep learning, particularly Convolutional Neural Networks (CNNs), have transformed nuclei segmentation. This study emphasizes feature extraction, introducing the ConvNext Mixer‐based Encoder‐Decoder (CNM‐ED) model. Unlike traditional CNN based models, the proposed CNM‐ED model enables the extraction of spatial and long context features to address the inherent complexities of histopathology images. This method leverages a multi‐path strategy using a traditional CNN architecture as well as different paths focused on obtaining customized long context features using the ConvNext Mixer block structure that combines ConvMixer and ConvNext blocks. The fusion of these diverse features in the final segmentation output enables improved accuracy and performance, surpassing existing state‐of‐the‐art segmentation models. Moreover, our multi‐level feature extraction strategy is more effective than models using self‐attention mechanisms such as SwinUnet and TransUnet, which have been frequently used in recent years. Experimental studies were conducted using five different datasets (TNBC, MoNuSeg, CoNSeP, CPM17, and CryoNuSeg) to analyze the performance of the proposed CNM‐ED model. Comparisons were made with various CNN based models in the literature using evaluation metrics such as accuracy, AJI, macro F1 score, macro intersection over union, macro precision, and macro recall. It was observed that the proposed CNM‐ED model achieved highly successful results across all metrics. Through comparisons with state‐art‐of models from the literature, the proposed CNM‐ED model stands out as a promising advancement in nuclei segmentation, addressing the intricacies of histopathological images. The model demonstrates enhanced diagnostic capabilities and holds the potential for significant progress in medical research. [ABSTRACT FROM AUTHOR]

In recent years, many studies have been carried out in the field of artificial intelligence in the literature with the development of equipment. Face recognition algorithms have an important place among these developments. Among the face recognition algorithms, the most successful ones are usually deep learning approaches. Models such as SphereFace, CosFace, ArcFace, and MagFace are important deep learning models in the literature. Despite their success, deep learning models are often computationally costly. Therefore, advanced methods are needed to reduce the computational load for these models. One of the most valid methods for this is to choose the most valuable one among embedding features for face recognition. Thus, cost can be reduced, and accuracy values can be increased even more. In this study, the most valuable of the 512 embedded features in the MagFace model was tried to be obtained by using PSO, GA, SCA, and DE optimization algorithms. As a result, accuracy values of 99.83%, 98.57%, and 98.65% were reached for 193, 252, and 280 features selected in the LFW, CFP, and AGEDB datasets, respectively. [ABSTRACT FROM AUTHOR]

Retinal blood vessels are directly or indirectly associated with many diseases. The retinal blood vessels consist of artery and vein vessels. With the automatic correct identification of these vessels, many diseases can be prevented. In this paper, a method is proposed to separate between arteries and veins on retinal blood vessel images. In the proposed method, firstly, the image preprocessing step is applied. Then, image patches are obtained from pre-processed retinal fundus images. These patches are prepared as input to the deep learning network architecture. The proposed deep learning network architecture is presented as a new CNN architecture. This architecture decides whether the blood vessel pixels in the fundus image are arteries or veins. The proposed method was evaluated on publicly available given DRIVE, INSPIRE datasets, and the recently created LES-AV dataset. The performance of the proposed method was evaluated by using the most commonly used sensitivity, specificity, and accuracy performance measures. The accuracy measure for all vessel pixels is 0.9110 for DRIVE, 0.9654 for INSPIRE, and 0.9531 for LES-AV dataset. The proposed method is compared with other state-of-the-art artery/vein separation methods. The experimental results of the proposed method are promising. This method is suitable for automatic artery/vein separation. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Polytechnic is the property of Journal of Polytechnic and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

English is accepted as an academic language in the world. This necessitates the use of English in their academic studies for speakers of other languages. Even when these researchers are competent in the use of the English language, some mistakes may occur while writing an academic article. To solve this problem, academicians tend to use automatic translation programs or get assistance from people with an advanced level of English. This study offers an expert system to enable assistance to the researchers throughout their academic article writing process. In this study, Turkish which is considered among low-resource languages is used as the source language. The proposed model combines the transformer encoder-decoder architecture model with the pre-trained Sci-BERT language model via the shallow fusion method. The model uses a Fully Attentional Network Layer instead of a Feed-Forward Network Layer in the known shallow fusion method. In this way, a higher success rate could be achieved by increasing the attention at the word level. Different metrics were used to evaluate the created model. The model created as a result of the experiments reached 45.1 BLEU and 73.2 METEOR scores. In addition, the proposed model achieved 20.12 and 20.56 scores, respectively, with the zero-shot translation method in the World Machine Translation (2017–2018) test datasets. The proposed method could inspire other low-resource languages to include the language model in the translation system. In this study, a corpus composed entirely of academic sentences is also introduced to be used in the translation system. The corpus consists of 1.2 million parallel sentences. The proposed model and corpus are made available to researchers on our GitHub page. [ABSTRACT FROM AUTHOR]

Copyright of Gazi Üniversitesi Fen Bilimleri Dergisi Part C: Tasarım ve Teknoloji is the property of Gazi University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Image processing-based fire/flame detection has become popular in recent years. In this paper, a novel fire/flame detection system based on a new conversion matrix and artificial bee colony algorithm was presented. Flame and non-flame image pixel values were combined to have a new feature matrix. A conversion matrix was generated randomly. The conversion matrix was multiplied by the feature matrix. The error of this multiplication result was calculated using the K-means clustering algorithm. The conversion matrix was updated until getting desired performance using artificial bee colony algorithm. At the end of the updating process, updated conversion matrix was multiplied with all images in the dataset to move all images to new color space. The final images were converted into binary images. Otsu method was used to get binary images. These binary images were compared with the corresponding ground truth images in the dataset. The aim of this comparison is to calculate the similarity ratio of the two images. This ratio shows the extent to which the original image features are preserved. A forest fire dataset was used which has 500 forest fire images. It is publicly available and called as Corsican Fire Database. Jaccard and Dice similarity measure parameters were used to evaluate the proposed system performance and compared with other similar study such as particle swarm optimization. Evaluated mean Jaccard index value was 0.76, and mean Dice index value was 0.85. This evaluation was made for 500 images. These results provide that this system can be used in fire/flame detection systems. [ABSTRACT FROM AUTHOR]

The timely and accurate diagnosis of plant diseases plays an important role in preventing the loss of productivity and loss or reduced quantity of agricultural products. In order to solve such problems, methods based on machine learning can be used. In recent years, deep learning, which is especially widely used in image processing, offers many new applications related to precision agriculture. In this study, we evaluated the performance results using different approaches of nine powerful architectures of deep neural networks for plant disease detection. Transfer learning and deep feature extraction methods are used, which adapt these deep learning models to the problem at hand. The utilized pretrained deep models are considered in the presented work for feature extraction and for further fine-tuning. The obtained features using deep feature extraction are then classified by support vector machine (SVM), extreme learning machine (ELM), and K-nearest neighbor (KNN) methods. The experiments are carried out using data consisting of real disease and pest images from Turkey. The accuracy, sensitivity, specificity, and F1-score are all calculated for performance evaluation. The evaluation results show that deep feature extraction and SVM/ELM classification produced better results than transfer learning. In addition, the fc6 layers of the AlexNet, VGG16, and VGG19 models produced better accuracy scores when compared to the other layers. [ABSTRACT FROM AUTHOR]

The brain cancer treatment process depends on the physician's experience and knowledge. For this reason, using an automated tumor detection system is extremely important to aid radiologists and physicians to detect brain tumors. The proposed method has three stages, which are preprocessing, the extreme learning machine local receptive fields (ELM-LRF) based tumor classification, and image processing based tumor region extraction. At first, nonlocal means and local smoothing methods were used to remove possible noises. In the second stage, cranial magnetic resonance (MR) images were classified as benign or malignant by using ELM-LRF. In the third stage, the tumors were segmented. The purpose of the study was using only cranial MR images, which have a mass, in order to save the physician's time. In the experimental studies the classification accuracy of cranial MR images is 97.18%. Evaluated results showed that the proposed method's performance was better than the other recent studies in the literature. Experimental results also proved that the proposed method is effective and can be used in computer aided brain tumor detection. [ABSTRACT FROM AUTHOR]