ing at the problem of low indoor visible light positioning accuracy, a positioning algorithm based on the received signal strength and arrival angle (RSS/AOA) information in indoor three-dimensional space is proposed. Based on the least squares (LS) criterion, the distance and angle are considered at the same time, a novel hybrid objective function is constructed, and then the least square estimator of location information is derived. For the non-convex objective function, the objective function is transformed into generalized trust region sub-problem (GTRS) to solve the global optimal solution. The simulation results show that in the 5 m×5 m×3 m two-dimensional positioning space, 20×20 test points are selected, for which the average positioning error is 8.7 cm. In addition, it can realize dynamic positioning and tracking in the three-dimensional space. The results show that the algorithm can obtain high accuracy in horizontal and vertical directions.

ing at the problems that traditional lung nodules detection methods can only get low sensitivities and a lot of false positives, this paper presents a retrieval method for lung nodules CT image based on end-to-end two-dimensional full convolution object recognition network (2D FCN) and three-dimensional target classification convolution neural network (3D CNN). Firstly, the method builds the 2D CNN for candidate selection to detect and locate the suspected regions on axial slices, and outputs an image that is the same size as the original image and is marked. Secondly, the three-dimensional patches of each candidate are extracted to train the 3D CNN. Finally, the trained 3D model is used to classify the false positive nodules. Experimental results on the LIDC-IDRI dataset show that the proposed method can achieve the recall rate of nodules of 98.2% at 36.2 false positives per scan. In the false positive reduction, the method respectively achieves high detection sensitivities of 87.3% and 97.0% at 1 and 4 false positives per scan. Experimental results on the LIDC-IDRI dataset show that the proposed method is highly suited to be used for lung nodules detection, achieves high recall rate and accuracy and outperforms the current reported method. Meanwhile, the proposed framework is general and can be easily extended to many other 3D object detection tasks from volumetric medical images, and it has an important application value in clinical practice with the aid of radiologists and surgeons.

of standard graphics library ALOI and many other real images indicate that the proposed algorithm improves the image matching accuracy and shortens the image matching time in the complex environment.

ing at the problems of non-contact images acquisition such as blur phenomenon, poor system identification systems and poor recognition effect, a palmprint and palm vein feature fusion recognition method based on block strengthened local directional pattern(BSLDP) and canonical correlation analysis is proposed. Firstly, we improve the traditional local directional pattern(LDP),and proposed the BSLDP algorithm to obtain the texture direction feature of palmprint and palm vein images. Secondly, the palmprint and palm vein feature fusion is realized effectively based on the canonical correlation analysis. Finally, the match identification is realized based on the chi-square distance. The experimental results show that the equal error rate is only 0.63% and 1.21% in the CASIA-M and the self-built non-contact image database. The results indicate that compared with other traditional and newest algorithms, the proposed method can eliminate redundant information, retain accurate feature information of palmprint and palm vein and improve system identification performance.