Manual crack inspection is labor-intensive and impractical at scale, prompting a shift toward AI-based segmentation methods. We present a novel crack segmentation model that leverages the Segment Anything Model 2 (SAM 2) through transfer learning to detect cracks on masonry surfaces. Unlike prior approaches that rely on encoders pretrained for image classification, we fine-tune SAM 2, originally trained for segmentation tasks, by freezing its Hiera encoder and FPN neck, while adapting its prompt encoder, LoRA matrices, and mask decoder for the crack segmentation task. No prompt input is used during training to avoid detection overhead. Our aim is to increase robustness to noise and enhance generalizability across different surface types. This work demonstrates the potential of foundational segmentation models in enabling more reliable and field-ready AI-based crack detection tools.
DOCUMENT
In this mixed methods study, a moderated mediation model predicting effects of leader-member exchange (LMX) and organizational citizenship behaviors (OCB) on innovative work behaviors, with employability as a mediator, has been tested. Multi-source data from 487 pairs of employees and supervisors working in 151 small and medium-sized enterprises (SMEs) supported our hypothesized model. The results of structural equation modelling provide support for our model. In particular, the benefits of close relationships and high-quality exchanges between employee and supervisor (LMX), and fostering individual development as a result of employees’ OCB have an indirect effect on innovative work behaviors through positive effects on workers’ employability. Innovative work behaviors depend on employees’ knowledge, skills, and expertise. In other words, enhancing workers’ employability nurtures innovative work behaviors. In addition, we found a moderation effect of organizational politics on the relationship between employability and innovative work behaviors. Secondly, qualitative methods focusing on experiences of the antecedents and outcomes of employability were used to complement our quantitative results. All in all, this study has important consequences for managerial strategies and practices in SMEs and call for an awareness of the dysfunctional effect of perceived organizational politics.
DOCUMENT
In the current market, the focus is more and more on building long-term relationships with clients in which value is created, communicated and delivered. This also means developing a new customer relationship focus that goes beyond the consumer orientation that has thus far dominated the practice of marketing. In a user-driven approach, the basis is no longer the product range, the world of the manufacturer or supplier, but rather the experience of the user. This has implications not only for manufacturers and service providers, but also for marketeers and market researchers. We feel that the most commonly used methods and techniques do not answer to the new demands of the market. We have therefore adapted emerging generative user research methods from the field of design as a basis for a truly user-driven market approach. As part of the Fitness in Motion project -a programme aimed at helping the health club industry develop a stronger customer orientation- we applied Contextmapping, a generative technique that has already proved its worth in the world of industrial product development and interaction design.
DOCUMENT
