Search Results (8)

Side-scan sonar images play an important role in underwater target detection, seabed mapping, and marine environment monitoring. However, the performance of deep learning-based detectors is often limited by the small scale of available sonar datasets, the high cost of data acquisition, and class imbalance among target categories. To address these issues, this paper proposes a diffusion-model-based data augmentation method for side-scan sonar target detection. A FLUX.1 diffusion model is adopted as the base generative framework and is fine-tuned using low-rank adaptation (LoRA) to adapt the pretrained model to the side-scan sonar image domain under limited training data conditions. The generated samples are further filtered and added only to the training set, while the validation and test sets are kept unchanged and contain only real sonar images. To ensure a fair evaluation of the augmentation strategy, all detection experiments are conducted using a fixed YOLOv8n (You Only Look Once version 8 nano) detector under the same training hyperparameters and three random seeds. Compared with training on the original dataset, the proposed FLUX+LoRA augmentation improves mean average precision (mAP)@0.5 from 0.7400 ± 0.0132 to 0.8582 ± 0.0328 and mAP@0.5:0.95 from 0.3994 ± 0.0187 to 0.5115 ± 0.0164. It also outperforms conventional augmentation methods under the same real-only validation/test protocol. In addition, Fréchet Inception Distance (FID)/Kernel Inception Distance (KID)-based image quality evaluation, generated-sample amount ablation, screening-strategy ablation, LoRA-rank sensitivity analysis, and a controlled 600-sample diffusion-backbone comparison are conducted. The results show that the 600-sample manually annotated FLUX+LoRA subset selected from generated samples achieves better image quality and detection performance than FLUX-base and SD1.5+LoRA under the same annotation budget. These findings demonstrate that FLUX+LoRA-generated sonar images can provide useful structural diversity for detector training and improve target detection performance under limited-data conditions. Full article

Reading proficiency in early childhood is crucial for academic success and intellectual development. However, more and more children are struggling with reading. According to the last PISA study in Austria, one out of five children is dealing with reading difficulties. The reasons for this are diverse, but an application that tracks children while reading aloud and guides them when they experience difficulties could offer meaningful help. Therefore, this proposal explores a prototyping approach for a core component that tracks children’s reading using a self-supervised Wav2Vec2 model with a limited amount of data. Self-supervised learning allows models to learn general representations from large amounts of unlabeled audio, which can then be fine-tuned on smaller, task-specific datasets, making it especially useful when labeled data is limited. Our model is operating on the phonetic level with the help of the International Phonetic Alphabet (IPA). To implement this, the KidsTALC dataset from the Leibniz University Hannover was used, which contains spontaneous speech recordings of German-speaking children. To enhance the training data and improve robustness, several data augmentation techniques were applied and evaluated, including pitch shifting, formant shifting, and speed variation. The models were trained using different data configurations to compare the effects of data variety and quality on recognition performance. The best model trained in this work achieved a phoneme error rate (PER) of 14.3% and a word error rate (WER) of 31.6% on unseen child speech data, demonstrating the potential of self-supervised models for such use cases. Full article

Unlike high-dose scans, low-dose cardiac CT perfusion imaging reduces patient radiation exposure and thereby the risk of potential health effects. However, it introduces significant image noise, degrading diagnostic quality and limiting clinical assessment. Denoising is thus a critical preprocessing step to enhance image quality without compromising anatomical or perfusion details. Traditionally used reconstruction-domain methods, such as Iterative Reconstruction and Compressed Sensing, are often limited by algorithmic complexity, dependence on raw sinogram data, and restricted adaptability. Conversely, image-domain methods offer more adaptable denoising options. Recently, learning-based approaches have further expanded this flexibility and demonstrated state-of-the-art performance across various denoising tasks. In this work, we present a deep learning-based denoising method specifically tuned for low-dose cardiac CT perfusion imaging. Our model is trained to reduce noise while preserving structural integrity and temporal contrast dynamics, which are critical for downstream analysis. Unlike many existing methods, our approach is optimized for perfusion data, where temporal consistency is essential. Residual cardiac motion remains a separate challenge, which we aim to address in our future work. Experimental results show significant improvements in quantitative image quality, using both reference-based and no-reference metrics, such as MSE/PSNR/SSIM and NIQE/FID/KID, as well as improved accuracy of perfusion measurements. Full article

Background: A growing body of literature examines the utility of emotion-focused parenting programs, as behaviorally based programs currently dominate the parenting literature. Few of those studies examine differences in how Black parents may benefit. This mixed-methods pilot study examined preliminary fidelity, efficacy, and acceptability of Tuning in to Kids (TIK), an emotion-focused parenting program targeting parenting practices and children’s emotion regulation through a strengths-based approach. Methods: Pre, post, and one-month follow-up measurements were collected from 21 parents in the United States who were randomly assigned to a treatment (i.e., TIK) or waitlist control group. They were assessed across several self-report parent measures (parental emotion regulation, emotion socialization parenting practices and beliefs) and parent-report of children’s social-emotional competence. Parents in the TIK group completed interviews to further understand their experience participating in the intervention. Results: Descriptive analyses showed general improvements and positive change in parenting practices, beliefs, parental emotion regulation, and children’s self-regulation. Large effect sizes indicate reductions of parents emotion dismissing and distressed reactions to children’s negative emotions. TIK was overall rated as a highly acceptable intervention. Parent interviews offer essential information to provide context to Black parents’ experiences utilizing TIK as well as themes related to challenges in raising Black children with self-regulation difficulties. Conclusions: Overall, these preliminary mixed-methods outcomes suggest that TIK is a promising parenting program to improve Black parents’ emotion regulation, emotion coaching beliefs and positive parenting practices. Further research is needed to investigate the effectiveness of TIK and other emotion-focused parenting programs with Black parents and assess the necessity of future cultural adaptations. Full article

Background: Parenting interventions based on emotion socialization (ES) theory offer an important theoretically driven approach to improve children’s emotional competence and behavioral functioning. Whether such approaches are effective in different cultural contexts, and whether the methods of delivery used are appropriate and acceptable, is an important empirical question. This paper reports on the preliminary evaluation of an ES parenting intervention, Tuning in to Kids (TIK), in Germany, Turkey, Iran, and China. Pilot studies of TIK have been conducted in each country with mothers of 4–6-year-old children. Method: The current study used qualitative methods with thematic analysis to explore the cultural appropriateness of the program in each site. Results: Culture-specific challenges were found across all sites in changing parents’ beliefs about the value of encouraging children’s emotional expression and supportive emotion discussions. Emotion literacy of parents depended on their access to emotion terms in their language, but also to parents’ experiences with emotions in their family of origin and culture-related beliefs about emotions. Adaptations were required to slow the speed of delivery, to address issues of trust with parents in seeking help, and to provide more opportunities to practice the skills and integrate different beliefs about parenting. Conclusion: While this ES parenting intervention has been developed in a Western cultural context, slight adaptations to the delivery methods (rather than change to the content) appeared to contribute to cultural appropriateness. The next step will be to quantitatively evaluate these adaptations of TIK in the different countries using randomized controlled studies. Full article

This study evaluated the Tuning in to Kids (TIK) parenting program delivered in a clinical setting with 77 parents and caregivers (hereafter referred to as “parents”) of children who had experienced complex trauma. The TIK program targets parent emotion socialization to improve children’s emotional and behavioral functioning. The study utilized a single-group design with pre- and post-intervention measures. Seventy-seven parents of children (aged 3–15 years) who had experienced complex trauma completed a ten-week version of the Trauma-Focused Tuning in to Kids program (TF-TIK). Measures examined parent reports of: emotion socialization; parent-child relationship; parent mental health; children’s emotional and behavioral functioning. Parents reported significantly improved emotion socialization, parent-child relationship, parent mental health, as well as child emotion regulation and behavior. This study provides initial support for the use of the TF-TIK parenting program in a clinical setting with parents of children who have experienced complex trauma in order to prevent or reduce problems. Full article

Movies have become one of the major sources of entertainment in the current era, which are based on diverse ideas. Action movies have received the most attention in last few years, which contain violent scenes, because it is one of the undesirable features for some individuals that is used to create charm and fantasy. However, these violent scenes have had a negative impact on kids, and they are not comfortable even for mature age people. The best way to stop under aged people from watching violent scenes in movies is to eliminate these scenes. In this paper, we proposed a violence detection scheme for movies that is comprised of three steps. First, the entire movie is segmented into shots, and then a representative frame from each shot is selected based on the level of saliency. Next, these selected frames are passed from a light-weight deep learning model, which is fine-tuned using a transfer learning approach to classify violence and non-violence shots in a movie. Finally, all the non-violence scenes are merged in a sequence to generate a violence-free movie that can be watched by children and as well violence paranoid people. The proposed model is evaluated on three violence benchmark datasets, and it is experimentally proved that the proposed scheme provides a fast and accurate detection of violent scenes in movies compared to the state-of-the-art methods. Full article

Air quality has had a significant impact on public health, the environment and eventually on the economy of countries for decades. Effectively mitigating air pollution in urban areas necessitates accurate air quality exposure information. Recent advancements in sensor technology and the increasing popularity of volunteered geographic information (VGI) open up new possibilities for air quality exposure assessment in cities. However, citizens and their sensors are put in areas deemed to be subjectively of interest (e.g., where citizens live, school of their kids or working spaces), and this leads to missed opportunities when it comes to optimal air quality exposure assessment. In addition, while the current literature on VGI has extensively discussed data quality and citizen engagement issues, few works, if any, offer techniques to fine-tune VGI contributions for an optimal air quality exposure assessment. This article presents and tests an approach to minimise land use regression prediction errors on citizen-contributed data. The approach was evaluated using a dataset (N = 116 sensors) from the city of Stuttgart, Germany. The comparison between the existing network design and the combination of locations selected by the optimisation method has shown a drop in spatial mean prediction error by 52%. The ideas presented in this article are useful for the systematic deployment of VGI air quality sensors, and can aid in the creation of higher resolution, more realistic maps for air quality monitoring in cities. Full article