Joint vehicle localization and categorization in high resolution aerial images can provide useful information for applications such as traffic flow structure analysis. To maintain sufficient features to recognize small-scaled vehicles, a regions with convolutional neural network features (R-CNN) -like detection structure is employed. In this setting, cascaded localization error can be averted by equally treating the negatives and differently typed positives as a multi-class classification task, but the problem of class-imbalance remains. To address this issue, a cost-effective network extension scheme is proposed. In it, the correlated convolution and connection costs during extension are reduced by feature map selection and bi-partite main-side network construction, which are realized with the assistance of a novel feature map class-importance measurement and a new class-imbalance sensitive main-side loss function. By using an image classification dataset established from a set of traditional real-colored aerial images with 0.13 m ground sampling distance which are taken from the height of 1000 m by an imaging system composed of non-metric cameras, the effectiveness of the proposed network extension is verified by comparing with its similarly shaped strong counter-parts. Experiments show an equivalent or better performance, while requiring the least parameter and memory overheads are required.

When compared with the traditional manual design method, the convolutional neural network has the advantages of strong expressive ability and it is insensitive to scale, light, and deformation, so it has become the mainstream method in the object detection field. In order to further improve the accuracy of existing object detection methods based on convolutional neural networks, this paper draws on the characteristics of the attention mechanism to model color priors. Firstly, it proposes a cognitive-driven color prior model to obtain the color prior features for the known types of target samples and the overall scene, respectively. Subsequently, the acquired color prior features and test image color features are adaptively weighted and competed to obtain prior-based saliency images. Finally, the obtained saliency images are treated as features maps and they are further fused with those extracted by the convolutional neural network to complete the subsequent object detection task. The proposed algorithm does not need training parameters, has strong generalization ability, and it is directly fused with convolutional neural network features at the feature extraction stage, thus has strong versatility. Experiments on the VOC2007 and VOC2012 benchmark data sets show that the utilization of cognitive-drive color priors can further improve the performance of existing object detection algorithms.

Existing image classification methods based on convolutional neural networks usually use a large number of samples to learn classification features hierarchically, causing the problems of over-fitting and error propagation layer by layer. Thus, they are vulnerable to adversarial samples generated by adding imperceptible disturbances to input samples. To address the above issue, we propose a cognitive-driven color prior model to memorize the color attributes of target samples inspired by the characteristics of human memory. At inference stage, color priors are indexed from the memory and fused with features of convolutional neural networks to achieve robust image classification. The proposed color prior model is cognitive-driven and has no training parameters, thus it has strong generalization and can effectively defend against adversarial samples. In addition, our method directly combines the features of the prior model with the classification probability of the convolutional neural network, without changing the network structure and its parameters of the existing algorithm. It can be combined with other adversarial attack defense methods, such as various preprocessing modules such as PixelDefense or adversarial training methods, to improve the robustness of image classification. Experiments on several benchmark datasets show that the proposed method improves the anti-interference ability of image classification algorithms.