In order to improve the accuracy and speed of railway track detection, a new method of detecting left and right railway tracks based on deep Convolutional Neural Network (CNN) and clustering was proposed. Firstly, the labeled images in the dataset were processed, each origin labeled image was divided into many grids uniformly, and the railway track information in each grid region was represented by one pixel, so as to construct the reduced images of railway track labeled images. Secondly, based on the reduced labeled images, a new deep CNN for railway track detection was proposed. Finally, a clustering method was proposed to distinguish left and right railway tracks. The proposed left and right railway track detection method can reach accuracy of 96% and speed of 155 frame/s on images with size of 1000 pixel×1000 pixel. Experimental results demonstrate that the proposed method not only has high detection accuracy, but also has fast detection speed.

Optimizing the color spraying sequence of auto parts can help enterprises to further reduce the production costs. However, there is no research proposing specific mathematical models and solutions for this type of problems. Considering the fact that each auto part must be sprayed and sprayed only once, which has the basic characteristics of the Traveling Salesman Problem (TSP), a modeling method of TSP transformation was proposed and the Genetic Algorithm (GA) with strong parallelism and robustness was selected to solve it. Firstly, the auto parts were defined as the TSP vertices, and the distance between the vertices and production constraints were defined according to the color and category requirements of the auto parts, so as to construct a 0-1 planning model for minimizing the number of color switching of the spraying sequence. Secondly, the color and category constraints of the auto parts were transformed into penalty factors, so as to construct the fitness function of the genetic algorithm. And, based on the championship selection strategy, the mutation strategies of copying, swapping, flipping, and sliding were comprehensively designed. Finally, three sets of data with 64, 93, 293 auto parts and 5, 7, 10 colors were constructed for simulation experiments. The proposed algorithm can obtain the accurate optimal solutions 5, 7, 10 for all three sets of data. With repeatedly running the algorithm, the average values of the approximate optimal solution are 5.63, 7.37, 11.49. Experimental results show that the established mathematical model is accurate to describe the auto part spraying sequence problem, and the designed genetic algorithm is efficient and practical, both of them can be applied to other similar production and processing problems.

In the diagnosis of rectal cancer, if the rectal tumor area can be automatically and accurately segmented from Computed Tomography (CT) images, it will help doctors make a more accurate and rapid diagnosis. Aiming at the problem of rectal tumor segmentation, an automatic segmentation method of rectal tumor based on improved U-Net model was proposed. Firstly, the sub coding modules were embedded in the U-Net model encoder of different levels to improve the feature extraction ability of the model. Secondly, by comparing the optimization performances of different optimizers, the most suitable optimizer was determined to train the model. Finally, data augmentation was performed to the training set to make the model more fully trained, so as to improve the segmentation performance. Experimental results show that compared with U-Net, Y-Net and FocusNetAlpha network models, the segmentation region obtained by the improved model is closer to the real tumor region, and the segmentation performance of this model for small objects is more prominent; at the same time, the proposed model is superior to other three models on three evaluation indexes including precision, recall and Dice coefficient, which can effectively segment the rectal tumor area.

Wrong object detection may lead to serious accidents, so high-precision object detection is very important in autonomous driving. An object detection method of Gaussian-YOLO v3 combining attention and feature intertwine module was proposed, in which several specific feature maps were mainly improved. First, the attention module was added to the feature map to learn the weight of each channel autonomously, enhancing the key features and suppressing the redundant features, so as to enhance the network ability to distinguish foreground object and background. Second, at the same time, different channels of the feature map were intertwined to obtain more representative features. Finally, the features obtained by the attention and feature intertwine modules were fused to form a new feature map. Experimental results show that the proposed method achieves mAP (mean Average Precision) of 20.81% and F ₁ score of 18.17% on BDD100K dataset, and has the false alarm rate decreased by 3.5 percentage points, reducing the false alarm rate effectively. It can be seen that the detection performance of the proposed method is better than those of YOLO v3 and Gaussian-YOLO v3.

Character recognition of license plate is an important component of an intelligent license plate recognition system. Both the number of categories and the complexity of background of license plate character affected the correct recognition rate. A character recognition method of license plate based on Convolution Neural Network (CNN) was proposed for improving the correct recognition rate. Firstly, the simple shape structures of license plate characters were obtained through image preprocessing which included image size normalization, image denoising, image binarization, image thinning, and character centering. Secondly, the preprocessed character images were trained and recognized by the proposed CNN model. The experimental results show that the correct recognition rate of the proposed method can reach 99.96%, which is better than the other three compared methods. It is demonstrated that the proposed CNN method has good recognition performance for the license plate character, and can meet the practical application requirements.