No abstract

No abstract

Service-based systems are a new software mode for distributed business processes integration. It is difficult for traditional testing methods to verify the functional and nonfunctional requirements of software. To address this problem, this paper proposes a visual verification platform to quantitatively compute the reliability and cost for evaluating the performance of service-based systems in the design phase. First, an extended automata model namely Probabilistic Reward Labeled Transition System (PRLTS) is proposed to formalize both the functional behaviors and nonfunctional features. Then, the formal language of probabilistic model checker PRISM is introduced to show the grammar of the target verification codes that we want to transform. Second, XML description tags of Business Process Execution Language (BPEL) is parsed to generate the functional behaviors using different kinds of transformation rules, based on which the probability matrix and reward concept are employed to denote the service’s reliability and cost, respectively. Third, the PRLTS model is turned into the input language of PRISM, where the graphic description language DOT of Graphviz is used as an intermediary to display system behaviors in a visual way. The model layout allows the designer to manually adjust the behaviors of the PRLTS model, where verification codes can be dynamically updated according to the changes in modified information. Fourth, to perform quantitative verification, the verification property in the form of the Probabilistic Computation Tree Logic (PCTL) formula can be automatically generated when the requirement model of the service-based system is inputted, during which the threshold value of qualitative property will be initially computed and returned as a recommended value. This allows the user to modify the qualitative property in an interactive way. Furthermore, experimental analysis of the real-world case study demonstrates the feasibility of the proposed method. Thus, our platform provides guidance for quantitative verification and graphical visualization for effectively generating formal models and checking the quantitative properties for service-based systems.

The surface defects’ region of strip steel is small, and has various defect types and, complex gray structures. There tend to be a large number of false defects and edge light interference, which lead traditional machine vision algorithms to be unable to detect defects for various types of strip steel. Image detection techniques based on deep learning require a large number of images to train a network. However, for a dataset with few samples with category imbalanced defects, common deep learning neural network training tasks cannot be carried out. Based on rapid image preprocessing algorithms (improved gray projection algorithm, ROI image augmentation algorithm) and transfer learning theory, this paper proposes a set of processes for complete strip steel defect detection. These methods achieved surface rapid screening, defect feature extraction, sample dataset’s category balance, data augmentation, defect detection, and classification. Through verification of the mixed dataset, composed of the NEU surface dataset and dataset in this paper, the recognition accuracy of the improved VGG19 network in this paper reached 97.8%. The improved VGG19 network performs slightly better than the baseline VGG19 in six types of defects, but the improved VGG19 performs significantly better in the surface seams defects. The convergence speed and accuracy of the improved VGG19 network were taken into account, and the detection rate was greatly improved with few samples and imbalanced datasets. This paper also has practical value in terms of extending its method of strip steel defect detection to other products.

No abstract