Aiming at the problems of low efficiency， low accuracy， excessive occupancy of human resources and intelligent classification algorithm miniaturization deployment requirements in China Customs risk control methods at this stage， a customs risk control method based on an improved Butterfly Feedback neural Network Version 2 （BFNet-V2） was proposed. Firstly， the Filling in Code （FC） algorithm was used to realize the semantic replacement of the customs tabular data to the analog image. Then， the analog image data was trained by using the BFNet-V2. The regular neural network structure was composed of left and right links， different convolution kernels and blocks， and small block design， and the residual short path was added to improve the overfitting and gradient disappearance. Finally， a Historical momentum Adaptive moment estimation algorithm （H-Adam） was proposed to optimize the gradient descent process and achieve a better adaptive learning rate adjustment， and classify customs data. Xception （eXtreme inception）， Mobile Network （MobileNet）， Residual Network （ResNet）， and Butterfly Feedback neural Network （BF-Net） were selected as the baseline network structures for comparison. The Receiver Operating Characteristic curve （ROC） and the Precision-Recall curve （PR） of the BFNet-V2 contain the curves of the baseline network structures. Taking Transfer Learning （TL） as an example， compared with the four baseline network structures， the classification accuracy of BFNet-V2 increases by 4.30%，4.34%，4.10% and 0.37% respectively. In the process of classifying real-label data， the misjudgment rate of BFNet-V2 reduces by 70.09%，57.98%，58.36% and 10.70%， respectively. The proposed method was compared with eight classification methods including shallow and deep learning methods， and the accuracies on three datasets increase by more than 1.33%. The proposed method can realize automatic classification of tabular data and improve the efficiency and accuracy of customs risk control.

Aiming at convex optimization problem of undersampling image reconstruction, a new image reconstruction algorithm based on the second order Total Generalized Variation (TGV) model was proposed. In the new model, the second-order TGV semi-norm of images was used as the regularization term, which could automatically balance the first order and second order derivative. The characteristics of the TGV made the new model recover the image edge information better, smooth noise and avoid the staircasing effect. For computing the new model effectively, the orthogonal projection and the adjustment of weight threshold were presented to adaptively amend the iteration results of each step in order to obtain accurate image reconstruction results. The experimental results show that the proposed model can get better results with large value of Peak Signal-to-Noise Ratio (PSNR) and Structure SIMilarity (SSIM) in image reconstruction compared with Orthogonal Matching Pursuit (OMP) and Total Variation (TV) models.

With reference to the creation of the independent game StarMusiX, this paper provided a new method of designing game scenes in order to solve such problems as high cost and low efficiency in development and production of game scenes. The method is that external data was analyzed in real-time and the result was recognized as the driving factors of real-time game scenes building, meanwhile the details of game scenes were generated by program. In the experimental game, scenes can be generated by inputting and analyzing "music-data" and a full expression can be found in "music-visualization". This verifies the rationality and feasibility of the above methods. The experimental results indicate that the method effectively improves the efficiency of game scenes design.

A new matching selection called similarity-matching selection of genetic algorithm was presented and the probabilities of the selection were calculated.An experimental calculation based on the proposed matching selection for a maximum problem was performed.The results show that such a matching selection guarantees the centralization and continuity of the excellent genes and can help to maintain the good gene constructions from the point of real world’s view.Furthermore,from the point of calculation convergence view,the calculation using the new matching selection is not easy to diverge in the small area around the global maximum.Therefore,the speed of convergence can be accelerated.