HMD-Net: A Vehicle Hazmat Marker Detection Benchmark

Vehicles carrying hazardous material (hazmat) are severe threats to the safety of highway transportation, and a model that can automatically recognize hazmat markers installed or attached on vehicles is essential for intelligent management systems. However, there is still no public dataset for benchmarking the task of hazmat marker detection. To this end, this paper releases a large-scale vehicle hazmat marker dataset named VisInt-VHM, which includes 10,000 images with a total of 20,023 hazmat markers captured under different environmental conditions from a real-world highway. Meanwhile, we provide an compact hazmat marker detection network named HMD-Net, which utilizes a revised lightweight backbone and is further compressed by channel pruning. As a consequence, the trained-model can be efficiently deployed on a resource-restricted edge device. Experimental results demonstrate that compared with some established methods such as YOLOv3, YOLOv4, their lightweight versions and popular lightweight models, HMD-Net can achieve a better trade-off between the detection accuracy and the inference speed.     

基于MobileNet-SSD的红外人脸检测算法

针对"可见光"+"热红外"融合的双光人脸检测算法对硬件依赖性高的劣势.本文提出一种基于MobileNet-SSD的红外人脸检测算法,该算法可以直接检测出红外图像中的人脸区域,对硬件的依赖性较低.同时实验结果表明该算法的检测精度和实时性均有所提高,可以直接运用到"疫情"期间的智能体温监控系统中.     

基于TensorFlow Lite平台的实时目标检测

目标检测是计算视觉的重要研究方向之一,尤其是基于移动设备平台实现快速精准的目标检测功能是非常有必要的.为了能够在移动设备上进行实时目标检测,本文提出一种基于Raspberry Pi 4B硬件平台,采用TensorFlow Lite开发环境,加载MobileNet-SSD网络结构算法的方案.方案采用的MobileNet卷...     

一种基于改进的MobileNetV2网络语义分割算法

基于金字塔卷积神经网络的语义分割算法准确率很高,但是其计算资源消耗巨大、算法执行时间长、无法满足实时性要求.为了解决这个问题,本文做出了以下改进：（1）用MobileNet替换原网络的结构,减少了网络运算时间和内存开销;（2）引入编码器-解码器结构提高输出图像的分辨率,进一步细化分割结果;（3）针对高分辨率图像推断时间过长的问题,本文设计了多级图像输入方法,降低了网络推断高分辨率图像所消耗的时间.本文在VOC 2012数据集和Cityscapes数据集上进行了测试,并与FCN、SegNet、DeepLab、PSPNet以及DFN等语义分割模型对比.实验结果表明,本文设计的语义分割算法在VOC 2012数据集上达到了76.1%的mIoU,在Cityscapes数据集上达到了74.1%的mIoU,略低于传统语义分割算法;处理一张分辨率为1024×512的图片需要18ms,少于传统语义分割算法,满足了实时性要求,达到了准确率与计算资源消耗之间的平衡.     

基于改进SSD的水下光学图像感兴趣目标检测算法研究

针对轻量化目标模型SSD-MV2对水下光学图像感兴趣目标检测精度低的问题,该文提出一种通道可选择的轻量化特征提取模块(SEB)和一种卷积核可变形、通道可选择的特征提取模块(SDB)。与此同时,利用SEB模块和SDB模块分别重新设计了SSD-MV2的基础网络和附加特征提取网络,记作SSD-MV2SDB,并为其选择了合理的基础网络扩张系数和附加特征提取网络SDB模块数量。在水下图像感兴趣目标检测数据集UOI-DET上,SSD-MV2SDB比SSD-MV2检测精度提高3.04%。实验结果表明,SSD-MV2SDB适用于水下图像感兴趣目标检测任务。     

A Lightweight YOLOv4-Based Forestry Pest Detection Method Using Coordinate Attention and Feature Fusion

Traditional pest detection methods are challenging to use in complex forestry environments due to their low accuracy and speed. To address this issue, this paper proposes the YOLOv4_MF model. The YOLOv4_MF model utilizes MobileNetv2 as the feature extraction block and replaces the traditional convolution with depth-wise separated convolution to reduce the model parameters. In addition, the coordinate attention mechanism was embedded in MobileNetv2 to enhance feature information. A symmetric structure consisting of a three-layer spatial pyramid pool is presented, and an improved feature fusion structure was designed to fuse the target information. For the loss function, focal loss was used instead of cross-entropy loss to enhance the network’s learning of small targets. The experimental results showed that the YOLOv4_MF model has 4.24% higher mAP, 4.37% higher precision, and 6.68% higher recall than the YOLOv4 model. The size of the proposed model was reduced to 1/6 of that of YOLOv4. Moreover, the proposed algorithm achieved 38.62% mAP with respect to some state-of-the-art algorithms on the COCO dataset.     

基于改进SSD的合成孔径声呐图像水下多尺度目标轻量化检测模型

针对轻量化目标检测模型SSD-MV2对合成孔径声呐(SAS)图像水下多尺度目标检测精度低的问题,该文提出一种新的卷积核模块-可扩张可选择模块(ESK),ESK具有通道可扩张、通道可选择和模型参数少的优点。与此同时,利用ESK模块重新设计了SSD的基础网络和附加特征提取网络,记作SSD-MV2ESK,并为其选择了合理的扩张系数和多尺度系数。在合成孔径声呐图像水下多尺度目标检测数据集SST-DET上,SSD-MV2ESK在模型参数基本相等的条件下,检测精度比SSD-MV2提升4.71%。实验结果表明,SSD-MV2ESK适用于合成孔径声呐图像水下多尺度目标检测任务。     

A novel MobileNet with selective depth multiplier to compromise complexity and accuracy

In the last few years, convolutional neural networks (CNNs) have demonstrated good performance while solving various computer vision problems. However, since CNNs exhibit high computational complexity, signal processing is performed on the server side. To reduce the computational complexity of CNNs for edge computing, a lightweight algorithm, such as a MobileNet, is proposed. Although MobileNet is lighter than other CNN models, it commonly achieves lower classification accuracy. Hence, to find a balance between complexity and accuracy, additional hyperparameters for adjusting the size of the model have recently been proposed. However, significantly increasing the number of parameters makes models dense and unsuitable for devices with limited computational resources. In this study, we propose a novel MobileNet architecture, in which the number of parameters is adaptively increased according to the importance of feature maps. We show that our proposed network achieves better classification accuracy with fewer parameters than the conventional MobileNet.