Search

Triplet network is an efficient method of metric learning, but with the increase of the number of fine-grained images and sample categories, the training of Triplet network is more and more challengeable. In order to solve this problem, this paper proposes an algorithm that effectively combine Concept Ontology Structure with the Triplet network trained of Two-layer Ontology Loss. It not only utilizes semantic knowledge to guide the Concept Ontology Structure of the network, but also makes use of the relationship between the layers to make the network more effective to see the triplets, which enhances the separability of the learned features. At the same time, we also use the bilinear function jointly trained with the Triplet network to enhance the image details, further improving the performance of the network. Finally, the effectiveness of the proposed algorithm is also proved by the results of classification experiments on the fine-grained image databases - Orchid and Fashion60.

Airport objects are hotspots in the field of image object detection because of their specific features and value for applications. In this study, we developed a complex object detection method based on improved Faster R-CNN to achieve higher detection precision to detect seven types of remote sensing image objects in airport areas under complex conditions such as different scales, different visual angles, and different backgrounds. When building the network, we used deeper basic networks and feature fusion components to extract more robust features. At the same time, we had also modified the selection of positive and negative samples to improve sample imbalance. The main improvements in the algorithm concern the anchor size generation rule, and the addition of an a priori judgment network for the network. The effectiveness of the improved algorithm was verified in experiments. Compared with the original Faster R-CNN, the improved network brings a 12.7% increase in mAP, at the detection time of 0.307s. Finally, the model with trained weights was used to test the detection of the seven types of objects in airport areas on different data sets, and comparisons were conducted with other algorithms. The experimental results showed that the method improved the average detection accuracy and had a good performance in remote sensing airport object detection tasks.

In this paper, a novel hierarchical deep network is proposed to combine the deep convolutional neural network and the feature selection-based tree classifier efficiently for image classification. First, the concept ontology is built for organizing large-scale image classes hierarchically in a coarse-to-fine fashion. Second, a novel selective orthogonal algorithm is proposed to make sure deep features extracted for each level classifiers more in line with the requirements of different classification tasks. Also, the role of useful feature components in multi-level deep features are improved. The experimental results on three datasets show that adding a feature selection module in a hierarchical deep network can perform better performance in large-scale image classification.

Aiming at the problem of insufficient representation ability of weak and small objects and overlapping detection boxes in airplane object detection, an effective airplane detection method in remote sensing images based on multilayer feature fusion and an improved nonmaximal suppression algorithm is proposed. Firstly, based on the common low-level visual features of natural images and airport remote sensing images, region-based convolutional neural networks are chosen to conduct transfer learning for airplane images using a limited amount of data. Then, the L2 norm normalization, feature connection, scale scaling, and feature dimension reduction are introduced to achieve effective fusion of low- and high-level features. Finally, a nonmaximal suppression method based on a soft decision function is proposed to solve the overlap problem of detection boxes. The experimental results show that the proposed method can effectively improve the representation ability of weak and small objects, as well as quickly and accurately detect airplane objects in the airport area.

Fast and accurate airport detection in remote sensing images is important for many military and civilian applications. However, traditional airport detection methods have low detection rates, high false alarm rates and slow speeds. Due to the power convolutional neural networks in object-detection systems, an end-to-end airport detection method based on convolutional neural networks is proposed in this study. First, based on the common low-level visual features of natural images and airport remote sensing images, region-based convolutional neural networks are chosen to conduct transfer learning for airport images using a limited amount of data. Second, to further improve the detection rate and reduce the false alarm rate, the concepts of “divide and conquer” and “integral loss’’ are introduced to establish cascade region proposal networks and multi-threshold detection networks, respectively. Third, hard example mining is used to improve the object discrimination ability and the training efficiency of the network during sample training. Additionally, a cross-optimization strategy is employed to achieve convolution layer sharing between the cascade region proposal networks and the subsequent multi-threshold detection networks, and this approach significantly decreases the detection time. The results show that the method established in this study can accurately detect various types of airports in complex backgrounds with a higher detection rate, lower false alarm rate, and shorter detection time than existing airport detection methods.