The state-of-the-art deep neural networks have made a great breakthrough in remote sensing image classification. However, the heavy dependence on large-scale data sets limits the application of the deep learning to synthetic aperture radar (SAR) automatic target recognition (ATR) field where the target sample set is generally small. In this work, a metalearning framework named MSAR, consisting of a metalearner and a base-learner, is proposed to solve the sample restriction problem, which can learn a good initialization as well as a proper update strategy. After training, MSAR can implement fast adaptation with a few training images on new tasks. To the best of our knowledge, this is the first study to solve a few-shot SAR target classification via metalearning. In particular, the few-task problem is defined by analyzing the effect of available training classes on the performance of metalearning models. In order to reduce the metalearning difficulties caused by the few-task problem, three transfer-learning methods are employed, which can leverage the prior knowledge from the pretraining phase. Besides, we design a hard task mining method for effective metalearning. Based on the Moving and Stationary Target Acquisition and Recognition (MSTAR) data set, a specialized data set named NIST-SAR is devised to train and evaluate the proposed method. The experiments on NIST-SAR have shown that the proposed method yields better performances with the largest absolute improvements of 1.7% and 2.3% for 1-shot and 5-shot, respectively, over the next best, which indicates that the proposed method is promising and metalearning is a feasible solution for few-shot SAR ATR.
State-of-the-art object detection frameworks require the training on large-scale datasets, which is the crux of the present dilemma: overfitting or degrading performance with insufficient samples and time-consuming training process. On the basis of meta-learning, this paper proposes a generalized Few-Shot Detection (FSD) framework to overcome the above drawbacks of the current advances in object detection. The proposed framework consists of a meta-learner and an object detector. It can learn the general knowledge and proper fast adaptation strategies across many tasks. The meta-learner can teach the detector how to learn from few examples in just one updating step. Here, the object detector can be any supervised learning detection models in theory. Specifically, the proposed FSD framework employs Single-Shot MultiBox Detector (SSD) as the object detector in this work, thus called Meta-SSD. Besides, a novel benchmark is constructed from Pascal VOC dataset for training and evaluation of meta-learning FSD. Experiments show that Meta-SSD yields a promising result for FSD. Furthermore, the properties of Meta-SSD is analyzed. This work can serve as a strong baseline and provide some inspiration for meta-learning FSD.
Semantic segmentation performs pixel-wise classification for given images, which can be widely used in autonomous driving, robotics, medical diagnostics and etc. The recent advanced approaches have witnessed rapid progress in semantic segmentation. However, these supervised learning based methods rely heavily on large-scale datasets to acquire strong generalizing ability, such that they are coupled with some constraints. Firstly, human annotation of pixel-level segmentation masks is laborious and time-consuming, which causes relatively expensive training data and make it hard to deal with urgent tasks in dynamic environment. Secondly, the outstanding performance of the above data-hungry methods will decrease with few available training examples. In order to overcome the limitations of the supervised learning semantic segmentation methods, this paper proposes a generalized meta-learning framework, named Meta-Seg. It consists of a meta-learner and a base-learner. Specifically, the meta-learner learns a good initialization and a parameter update strategy from a distribution of few-shot semantic segmentation tasks. The base-learner can be any semantic segmentation models theoretically and can implement fast adaptation (that is updating parameters with few iterations) under the guidance of the meta-learner. In this work, the successful semantic segmentation model FCN8s is integrated into Meta-Seg. Experiments on the famous few-shot semantic segmentation dataset PASCAL5i prove Meta-Seg is a promising framework for few-shot semantic segmentation. Besides, this method can provide with reference for the relevant researches of meta-learning semantic segmentation.
Synthetic Aperture Radar (SAR) target classification is an important branch of SAR image interpretation. The deep learning based SAR target classification algorithms have made remarkable achievements. But the acquisition and annotation of SAR target images are time-consuming and laborious, and it is difficult to obtain sufficient training data in many cases. The insufficient training data can make deep learning based models suffering from over-fitting, which will severely limit their wide application in SAR target classification. Motivated by the above problem, this paper employs transfer-learning to transfer the prior knowledge learned from a simulated SAR dataset to a real SAR dataset. To overcome the sample restriction problem caused by the poor feature discriminability for real SAR data. A simple and effective sample spectral regularization method is proposed, which can regularize the singular values of each SAR image feature to improve the feature discriminability. Based on the proposed regularization method, we design a transfer-learning pipeline to leverage the simulated SAR data as well as acquire better feature discriminability. The experimental results indicate that the proposed method is feasible for the sample restriction problem in SAR target classification. Furthermore, the proposed method can improve the classification accuracy when relatively sufficient training data is available, and it can be plugged into any convolutional neural network (CNN) based SAR classification models.
Short-term load in power system is nonlinear and non-stationary. To cope with the problem that the training error of neural network prediction model is increased while the generalization ability is reduced caused by the large input fluctuation, the rough neurons with upper and lower inputs are introduced into the radial basis function (RBF) neural networks, a power system short-term load forecasting model based on rough RBF neural networks is presented. The genetic algorithm is used to optimize related parameters of rough neural networks to find the best parameters in the global. The simulation results show that the proposed method can effectively improve the prediction accuracy of the power system short-term load.
Object detection has been a building block in computer vision. Though considerable progress has been made, there still exist challenges for objects with small size, arbitrary direction, and dense distribution. Apart from natural images, such issues are especially pronounced for aerial images of great importance. This paper presents a novel multi-category rotation detector for small, cluttered and rotated objects, namely SCRDet. Specifically, a sampling fusion network is devised which fuses multi-layer feature with effective anchor sampling, to improve the sensitivity to small objects. Meanwhile, the supervised pixel attention network and the channel attention network are jointly explored for small and cluttered object detection by suppressing the noise and highlighting the objects feature. For more accurate rotation estimation, the IoU constant factor is added to the smooth L1 loss to address the boundary problem for the rotating bounding box. Extensive experiments on two remote sensing public datasets DOTA, NWPU VHR-10 as well as natural image datasets COCO, VOC2007 and scene text data ICDAR2015 show the state-of-the-art performance of our detector. The code and models will be available at https://github.com/DetectionTeamUCAS.