当前位置: X-MOL 学术Pattern Recognit. Image Anal. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Memory Consumption and Computation Efficiency Improvements of Viola–Jones Object Detection Method for Remote Sensing Applications
Pattern Recognition and Image Analysis ( IF 0.7 ) Pub Date : 2021-09-21 , DOI: 10.1134/s1054661821030238
S. A. Usilin 1, 2 , O. A. Slavin 1, 2 , V. V. Arlazarov 1, 2
Affiliation  

Abstract

In this paper, we consider object classification and detection problems. We propose an algorithm that is effective from the point of view of computational complexity and memory consumption. The proposed algorithm can be successfully used as a basic tool for building different remote sensing systems which are, in general, installed on UAVs. The algorithm is based on the Viola–Jones method. It is shown in the paper, that the Viola–Jones method is the most preferable approach to detect objects on-board UAVs, because it needs the least amount of memory and the number of computational operations to solve the object detection problem. To ensure sufficient accuracy, we use a modified feature: rectangular Haar-like features, calculated over the magnitude of the image gradient. To increase computational efficiency, the L1 norm was used to calculate the magnitude of the image gradient. To train orientation-independent complex classifier we use a more generic decision tree form of complex classifier instead of a cascade scheme. All mentioned improvements were evaluated during detection of the following objects: the PSN-10 inflatable life raft (an example of an object that is detected during rescue operations using UAVs), oil tank storage (such kind of objects are usually detected during the inspection of industrial infrastructure), and aircraft on an area of hardstand. The performance of the trained detectors was estimated on real data (including data obtained during the rescue operation of the trawler Dalniy Vostok and a subset of real images from the DOTA dataset).



中文翻译:

用于遥感应用的 Viola-Jones 对象检测方法的内存消耗和计算效率改进

摘要

在本文中,我们考虑对象分类和检测问题。我们提出了一种从计算复杂度和内存消耗的角度来看有效的算法。所提出的算法可以成功地用作构建通常安装在无人机上的不同遥感系统的基本工具。该算法基于 Viola-Jones 方法。论文中表明,Viola-Jones 方法是检测机载无人机目标的最优选方法,因为它需要最少的内存和计算操作的数量来解决目标检测问题。为了确保足够的准确性,我们使用了一个修改过的特征:矩形 Haar-like 特征,在图像梯度的大小上计算。为了提高计算效率,L1范数用于计算图像梯度的大小。为了训练与方向无关的复杂分类器,我们使用更通用的复杂分类器的决策树形式而不是级联方案。所有提到的改进都在检测以下物体时进行了评估:PSN-10 充气救生筏(使用无人机在救援行动中检测到的物体的一个例子)、油罐储存(此类物体通常在检查过程中检测到)工业基础设施)和飞机在硬架区域。受训探测器的性能是根据真实数据(包括拖网渔船救援行动期间获得的数据)估算的 为了训练与方向无关的复杂分类器,我们使用更通用的复杂分类器的决策树形式而不是级联方案。所有提到的改进都在检测以下物体时进行了评估:PSN-10 充气救生筏(使用无人机在救援行动中检测到的物体的一个例子)、油罐储存(此类物体通常在检查过程中检测到)工业基础设施)和飞机在硬架区域。受训探测器的性能是根据真实数据(包括拖网渔船救援行动期间获得的数据)估算的 为了训练与方向无关的复杂分类器,我们使用更通用的复杂分类器的决策树形式而不是级联方案。所有提到的改进都在检测以下物体时进行了评估:PSN-10 充气救生筏(使用无人机在救援行动中检测到的物体的一个例子)、油罐储存(此类物体通常在检查过程中检测到)工业基础设施)和飞机在硬架区域。受训探测器的性能是根据真实数据(包括拖网渔船救援行动期间获得的数据)估算的 油罐存储(此类物体通常在工业基础设施检查期间检测到)和飞机在硬架区域。受训探测器的性能是根据真实数据(包括拖网渔船救援行动期间获得的数据)估算的 油罐存储(此类物体通常在工业基础设施检查期间检测到)和飞机在硬架区域。受训探测器的性能是根据真实数据(包括拖网渔船救援行动期间获得的数据)估算的Dalniy Vostok和来自 DOTA 数据集的真实图像子集)。

更新日期:2021-09-21
down
wechat
bug