基于Edge Boxes的大型车辆车标检测与识别

传统车标检测与识别算法难以检测大型车辆车标,且速度较慢。提出了一种基于Edge Boxes的大型车辆车标检测与识别方法。Edge Boxes算法是一种成熟的图像分割算法,能够快速且有效地检测物体位置,满足大型车辆车标检测与识别问题的准确性及实时性的需求。该方法首先根据车标在车辆中的空间位置关系初选车标候选区,然后利用Edge Boxes算法进行目标提取,进而将提取得到的目标送入利用线性约束编码构建的车标检测分类器和车标识别分类器进行训练与识别,得到车标检测与识别结果。对不同卡口的不同天气和光照条件下采集的4 480张图像（含50类大型车辆）进行实验,实验结果表明,在检测与识别性能以及时间消耗方面均优于传统方法,具有良好的实用前景。     

基于HOG特征和机器学习的工程车辆检测

针对智能监控在露天矿区中的应用, 提出了一种在大场景、远距离、多角度的环境下自动识别工程车辆的算法. 该算法利用梯度方向直方图特征(HOG)作为图像描述, 并结合支持向量机(SVM)训练分类器的方法, 实现图像中工程车辆的检测. 为提高检测效率和准确度, 采用Hough变换直线检测的方法提取包含目标的区域. 实验表明, 算法能够在较大分辨率的复杂图像中识别出检测目标, 有较高的准确度, 对于实验中不同的环境具有较强的鲁棒性.     

A real-time vehicle detection and a novel vehicle tracking systems for estimating and monitoring traffic flow on highways

Real-time highway traffic monitoring systems play a vital role in road traffic management, planning, and preventing frequent traffic jams, traffic rule violations, and fatal road accidents. These systems rely entirely on online traffic flow info estimated from time-dependent vehicle trajectories. Vehicle trajectories are extracted from vehicle detection and tracking data obtained by processing road-side camera images. General-purpose object detectors including Yolo, SSD, EfficientNet have been utilized extensively for real-time object detection task, but, in principle, Yolo is preferred because it provides a high frame per second (FPS) performance and robust object localization functionality. However, this algorithm’s average vehicle classification accuracy is below 57%, which is insufficient for traffic flow monitoring. This study proposes improving the vehicle classification accuracy of Yolo, and developing a novel bounding box (Bbox)-based vehicle tracking algorithm. For this purpose, a new vehicle dataset is prepared by annotating 7216 images with 123831 object patterns collected from highway videos. Nine machine learning-based classifiers and a CNN-based classifier were selected. Next, the classifiers were trained via the dataset. One out of ten classifiers with the highest accuracy was selected to combine to Yolo. This way, the classification accuracy of the Yolo-based vehicle detector was increased from 57% to 95.45%. Vehicle detector 1 (Yolo) and vehicle detector 2 (Yolo + best classifier), and the Kalman filter-based tracking as vehicle tracker 1 and the Bbox-based tracking as vehicle tracker 2 were applied to the categorical/total vehicle counting tasks on 4 highway videos. The vehicle counting results show that the vehicle counting accuracy of the developed approach (vehicle detector 2 + vehicle tracker 2) was improved by 13.25% and this method performed better than the other 3 vehicle counting systems implemented in this study.     

Face detection with boosted Gaussian features

Detecting faces in images is a key step in numerous computer vision applications, such as face recognition or facial expression analysis. Automatic face detection is a difficult task because of the large face intra-class variability which is due to the important influence of the environmental conditions on the face appearance. We propose new features based on anisotropic Gaussian filters for detecting frontal faces in complex images. The performances of our face detector based on these new features have been evaluated on reference test sets, and clearly show improvements compared to the state-of-the-art.     

一种改进的SSD网络车载图像检测方法

对行人和车辆的检测识别是无人驾驶领域的重要组成部分,为满足该领域对相关模型检测精确度的需求,以传统单发多框检测器(single shot multibox detector,SSD)为基础,提出了一种车载图像识别改进算法.鉴于传统SSD目标检测算法不能充分利用局部特征和全局语义特征、目标定位和识别存在矛盾等缺陷,提出了...     

基于神经网络的自然场景方向文本检测器

场景文本检测是场景文本识别中重要的一步,也是一个具有挑战性的问题。不同于一般的目标检测,场景文本检测的主要挑战在于自然场景图像中的文本具有任意方向,小的尺寸,以及多种宽高比。论文在TextBoxes[8]的基础上进行改进,提出了一个适用于任意方向文本的检测器,命名为OSTD(Oriented Scene Text Detector),可以有效且准确地检测自然场景中任意方向的文本。论文在公共数据集上对提出OSTD的进行评估。所有实验结果都表明,无论在准确性,还是实时性方面OSTD都是极具竞争力的方法。在1024×1024的ICDAR2015 Incidental Text数据集[16]上,OSTD的F-Measure=0.794,FPS=10.7。     

YOLOv3 Attention Face Detector with High Accuracy and Efficiency

In recent years, face detection has attracted much attention and achieved great progress due to its extensively practical applications in the field of face based computer vision. However, the tradeoff between accuracy and efficiency of the face detectors still needs to be further studied. In this paper, using Darknet-53 as backbone, we propose an improved YOLOv3-attention model by introducing attention mechanism and data augmentation to obtain the robust face detector with high accuracy and efficiency. The attention mechanism is introduced to enhance much higher discrimination of the deep features, and the trick of data augmentation is used in the training procedure to achieve higher detection accuracy without significantly affecting the inference speed. The model has been trained and evaluated on the popular and challenging face detection benchmark, i.e., the WIDER FACE training and validation subsets, respectively, achieving AP of 0.942, 0.919 and 0.821 with the speed of 28FPS. This performance exceeds some existing SOTA algorithms, demonstrating acceptable accuracy and near real time detection for VGA resolution images, even in the complex scenarios. In addition, the proposed model shows good generation ability on another public dataset FDDB. The results indicate the proposed model is a promising face detector with high efficiency and accuracy in the wild.     

基于选择性搜索算法的车脸部件检测

车脸部件检测能够广泛地应用于车辆识别及车辆的语义分割。尽管对于车脸的检测已经做出过大量的努力,但现有的研究大多集中在车脸的整块区域的检测及定位,提出了一种基于选择性搜索的车脸部件检测算法。该算法分为两个阶段：首先,将车辆图片进行高斯滤波去噪以及图像归一化预处理。其次,对预处理后的图片,利用基于图表示的图像分割算法获取初始分割区域,计算两两相邻区域在颜色、纹理、大小及吻合度之间的相似度;随后利用初始分割区域相邻区域间的颜色、纹理、大小以及吻合度的相似性对初始分割区域进行合并,从而准确分割车脸各部件。采用部分香港中文大学的公开数据集CompCars,总计4 199张图像,作为测试样本以测试车脸部件分割检测效果。实验结果表明,该算法检测车脸部件的平均重合度达到73.74%,明显胜过其它目标检测算法,此外,该算法不需训练,具有更强的通用性。     

YOLOv5定位多特征融合的车标识别

为解决智能交通系统中车标识别的问题,提出YOLOv5s网络车标定位多特征融合的车标图像识别方案。车标定位阶段选择YOLOv5s网络以满足对车标定位速度与精度等的需求。车标识别阶段通过调整扩展高斯差分中的参数得到具有不同效果的车标边缘,设计一组二维Gabor滤波器对边缘检测后的车标图像进行滤波处理并提取出对应的车标图像特征向量,通过计算待测车标图像特征与标准比对库中特征向量的欧几里德距离,取距离最小者对应的标签索引作为分类识别结果,该方案的最佳识别正确率为96.91%。采用随机森林算法进行分类后的最佳识别正确率可达99.33%。该方案的车标定位与识别最佳整体正确率超过了YOLOv5s网络直接一步到位识别车标的方案,且相较于传统图像处理方法有明显提升。     

交通道路行驶车辆车标识别算法

为解决交通道路行驶车辆车标识别中存在的目标小、噪声大、种类多的问题,提出了一种基于深度学习的目标检测算法与基于形态学模板匹配算法相结合的方法,并设计了一种高准确度且能应对新类型车标的识别系统.首先,采用通过K-Means++重新聚类锚框值,并引入残差网络的YOLOv4进行车标的一步定位;其次,通过对标准车标图像进行预处...     

复杂背景彩色图像中多角度人脸检测

提出一种针对复杂背景彩色图像中的人脸检测方法。基于Gray World假设,在RGB颜色空间采用颜色平衡的方法对偏色图像进行颜色校正,在YES颜色空间进行肤色检测并应用预处理技术缩小人脸检测的搜索区域。在物体区域方向计算的基础上,提出能够检测任意旋转角度人脸的方法,在人脸候选区域采用模板脸匹配方法定位人脸。实验表明,该方法对不同光照环境、不同尺寸、任意旋转的人脸有较好的检测效果。     

Contour detection improved by frequency domain filtering of gradient image

We propose an intermediate computational step,frequency domain filtering of gradient image,to improve contour detection performance of gradient-based edge detectors.This step is inspired by analyzing the spectrum distribution of object contours and texture edges in the frequency domain of gradient image.We illustrate the principle and efect of this step by adding it to the Canny edge detector.The resulting operator can selectively retain object contours and region boundaries,and meanwhile can dramatically reduce non-meaningful elements caused by textured background.We use several types of images to compare the proposed method and other related methods qualitatively and quantitatively.Experimental results show that the proposed method can efectively enhance the contour detection of Canny edge detector and achieves similar detection performance to two other related methods but runs faster.     

基于迁移学习的商品图像检测方法

近年来,对象识别方法被应用到多个领域.如人脸检测,车辆检测.然而模型训练所需要的边框标定需要很大的工作量.本文通过基于迁移学习的方法,将物体检测任务迁移到商品检测,且不需要边框标定.本文在分类层和边框回归层之间建立关系层,来学习两种任务之间的关联.本文建立了一个商品数据集,并提出了一种深度学习训练方法,解决了可旋转物体的检测问题.基于Faster RCNN框架,本文提出一种候选选择方法,可以在无边框标定情况下训练商品分类.本文提出的商品检测方法不需要边框标定,而且很容易训练并应用到其它数据集.     

Optimization of a training set for more robust face detection

The performance of a learning-based method highly depends on the quality of a training set. However, it is very challenging to collect an efficient and effective training set for training a good classifier, because of the high dimensionality of the feature space and the complexity of decision boundaries. In this research, we study the methodology of automatically obtaining an optimal training set for robust face detection by resampling the collected training set. We propose a genetic algorithm (GA) and manifold-based method to resample a given training set for more robust face detection. The motivations behind lie in two folds: (1) dynamic optimization, diversity, and consistency of the training samples are cultivated by the evolutionary nature of GA and (2) the desirable non-linearity of the training set is preserved by using the manifold-based resampling. We demonstrate the effectiveness of the proposed method through experiments and comparisons to other existing face detectors. The system trained from the training set by the proposed method has achieved 90.73% accuracy with no false alarm on MIT+CMU frontal face test set—the best result reported so far to our knowledge. Moreover, as a fully automatic technology, the proposed method can significantly facilitate the preparation of training sets for obtaining well-performed object detection systems in different applications.     

基于背景纹理的轿车车标定位方法

车标定位是车标识别系统的关键技术之一,但是由于车标背景的散热片纹理不一、种类繁多,给车标定位造成了困难,故提出了一种基于背景纹理的轿车车标定位方法.该方法首先根据先验知识对车标进行粗定位,依据其在水平投影与垂直投影上的特征将车标背景分为三大类,然后运用Sobel算子分别对不同类别的散热片背景进行消融;为了更好的去除散热片背景对定位车标的影响,引入了一种邻间二值化方法,同时结合基于投影的去噪方法对噪点进行进一步处理,从而实现车标的精确定位.这种方法适用于不同类型的车标背景条件下的车标定位.实验通过对1000张图片进行车标定位,比较已有算法有更高的准确率和适用性,总体定位准确率可以达到97.10%.     

辅助驾驶的车辆与行人检测模型YOLO-DNF

为了解决YOLO系列目标检测算法存在的精度与计算成本不均衡、模型泛化性不足的问题,提出了可满足不同光照场景下目标检测需求的高精度快速的车辆与行人检测模型YOLO-Day Night and Fast(YOLO-DNF)。文中结合当下主流检测模型所使用的卷积神经网络分析卷积结构与网络深度对于主干网络特征提取能力和计算成本的影响,针对网络不同层次选取卷积结构Arrow-Block与CSP-Block搭建网络并通过量化堆叠单元的计算成本确定网络深度,提出低计算成本、高特征提取能力的ACNet网络。此外分析了白天与夜间图像的亮度差异,引入了HSV域扰动并提出亮度处理的数据增强策略,提升了模型的夜间检测精度,改善了模型泛化性不足的问题。实验结果表明：YOLO-DNF模型在SODA10M数据集仅含白天图像的训练集中训练后以每秒24.36帧的检测速率达到32.8%的全时段mAP检测精度,检测精度与速度超过目前主流检测模型。其中夜间精度达到了27.7%,扩展了模型的检测应用场景。     

A data mining approach to face detection

In this paper, we propose a novel face detection method based on the MAFIA algorithm. Our proposed method consists of two phases, namely, training and detection. In the training phase, we first apply Sobel's edge detection operator, morphological operator, and thresholding to each training image, and transform it into an edge image. Next, we use the MAFIA algorithm to mine the maximal frequent patterns from those edge images and obtain the positive feature pattern. Similarly, we can obtain the negative feature pattern from the complements of edge images. Based on the feature patterns mined, we construct a face detector to prune non-face candidates. In the detection phase, we apply a sliding window to the testing image in different scales. For each sliding window, if the slide window passes the face detector, it is considered as a human face. The proposed method can automatically find the feature patterns that capture most of facial features. By using the feature patterns to construct a face detector, the proposed method is robust to races, illumination, and facial expressions. The experimental results show that the proposed method has outstanding performance in the MIT-CMU dataset and comparable performance in the BioID dataset in terms of false positive and detection rate.     

License plate detection based on multistage information fusion

Adaboost detector has been successfully used in object detection. In this paper, we propose a new License Plate (LP) detection technique based on multistage information fusion, which is adopted to reduce high false alarm rate in the conventional Adaboost detector. The proposed multistage information fusion system is composed of an enhanced Adaboost detector, a color checking module and an SVM detector, where the latter two stages further check whether the image patch that gets through the Adaboost detector is an LP. Test results of the dataset that consists of 950 real-world images show that the fusion reduces the false alarm rate. The proposed Fusion detector outperforms the conventional Adaboost detector throughout the ROC (Receiver Operating Characteristic) curve. The AUC (Area Under the Curve) of the best Fusion detector reaches 0.9081; however, the AUC of the best Adaboost detector is only 0.8441, which shows that the modification on feature extraction and the multistage information fusion significantly improve the LP detection performance.     

Moving shadow detection and removal for traffic sequences

Segmentation of moving objects in a video sequence is a basic task for application of computer vision. However, shadows extracted along with the objects can result in large errors in object localization and recognition. In this paper, we propose a method of moving shadow detection based on edge information, which can effectively detect the cast shadow of a moving vehicle in a traffic scene. Having confirmed shadows existing in a figure, we execute the shadow removal algorithm proposed in this paper to segment the shadow from the foreground. The shadow eliminating algorithm removes the boundary of the cast shadow and preserves object edges firstly; secondly, it reconstructs coarse object shapes based on the edge information of objects; and finally, it extracts the cast shadow by subtracting the moving object from the change detection mask and performs further processing. The proposed method has been further tested on images taken under different shadow orientations, vehicle colors and vehicle sizes, and the results have revealed that shadows can be successfully eliminated and thus good video segmentation can be obtained.