There are a variety of problems in the label printing production process, such as multi-variety, small batch, high-degree customization and uncertainties in some working procedures. Aiming at these problems, a flexible job-shop scheduling model with the goal of minimizing the maximum completion time was established, and an improved Genetic Algorithm (GA) was proposed. First of all, integer coding was adopted based on the standard genetic algorithm. Secondly, the roulette method was used in the selection operation stage, and the convergence of the algorithm was guaranteed by introducing the elite solution retention strategy. Finally, dynamic adaptive crossover and mutation probabilities were proposed to ensure that the algorithm optimized in a wide range to avoid prematurity in the early stage, and the algorithm converged timely to ensure that the excellent individuals obtained previously were not destroyed in the later stage. In order to verify the feasibility of the proposed improved genetic algorithm, the Ft06 benchmark example was first used to compare the proposed algorithm with the standard genetic algorithm. The results showed that the optimal solution of the improved genetic algorithm (55 s) was better than the optimal solution of the standard genetic algorithm (56 s), and the number of iterations of the improved genetic algorithm was significantly better than that of the standard genetic algorithm. Then, through the 8×8, 10×10 and 15×10 standard examples of Flexible Job-shop Scheduling Problem (FJSP), the effectiveness, stability and optimization performance of the algorithm were verified. On all of three standard test examples, the improved genetic algorithm obtained the optimal solution in a short time. Finally, when the proposed algorithm was used to solve the production scheduling problem of the label printing job-shop, the processing efficiency was increased by 50.3% compared to the original one. Therefore, the proposed improved genetic algorithm can be effectively applied to solve the production scheduling problem of label printing job-shop.

In order to utilize the shared information between adjacent versions in multi-version program verification, and extract and reuse loop invariants in the witness automaton belonging to the previous version, a software regression verification based on witness automata was proposed. Firstly, the witness file applicable to the new version of programs was generated by witness preprocessing. Then, based on the auxiliary-invariant-enhanced k-induction, the regression verification process was implemented to validate the new witness file and verify the new version of programs. Finally, performance of three kinds of regression verification was compared by contrast experiments, including the so-called "direct" verification that did not use invariant information and verification methods combined with or without data flow analysis. Compared with the direct verification, the time consumption of the regression verification combined with or without data flow analysis was reduced by 49% and 75% respectively, and the memory consumption was reduced by 18% and 50% respectively. The results show that when the program satisfies its verification specification, the regression verification based on witness automata can greatly improve verification efficiency, and the regression method combined with data flow analysis can make it even better.

Accurate background model is the paramount base for object extracting and tracing. In response to swing objects which part quasi-periodically changed in intricate scene, based on multi-Gaussian background model, a new Quasi-Periodic Background Algorithm (QPBA) was proposed to suppress the swing objects and establish an accurate and stable background model. The specific process included: According to multi-Gaussian background model, the object classification in scene was set up, and the effect on Gaussian model's parameters caused by swing objects was analyzed. By using color distribution values as samples to establish Gaussian model to keep swing pixels, the swing model in swing pixels was integrated into background model with weight factors of occurrence frequency and time interval. Comparison among QPBA and the classical background modeling algorithms such as GMM (Gaussian Mixture Model), ViBe (Visual Background extractor) and CodeBook was put forward, and the results were assessed in aspects of quality, quantity and efficiency. It shows that QPBA has a more obvious suppression on swing objects, and its fall-out ratio is less than 1%, so that it can handle the scene with swing objects. At the same time, its correct detection number is consistent with other algorithms, thus the moving objects can be reserved perfectly. In addition, the efficiency of QPBA is high, and its resolving time is approximate to CodeBook, which can satisfy the requirements of real-time computation.