Logformer: Cascaded Transformer for System Log Anomaly Detection

Modern large-scale enterprise systems produce large volumes of logs that record detailed system runtime status and key events at key points.These logs are valuable for analyzing performance issues and understanding the status of the system.Anomaly detection plays an important role in service management and system maintenance,and guarantees the reliability and security of online systems.Logs are universal semi-structured data,which causes difficulties for traditional manual detection and pattern-matching algorithms.While some deep learning algorithms utilize neural networks to detect anomalies,these approaches have an over-reliance on manually designed features,resulting in the effectiveness of anomaly detection depending on the quality of the features.At the same time,the aforementioned methods ignore the underlying contextual information present in adjacent log entries.We propose a novel model called Logformer with two cascaded transformer-based heads to capture latent contextual information from adjacent log entries,and leverage pre-trained embeddings based on logs to improve the representation of the embedding space.The proposed model achieves comparable results on HDFS and BGL datasets in terms of metric accuracy,recall and F1-score.Moreover,the consistent rise in F1-score proves that the representation of the embedding spacewith pre-trained embeddings is closer to the semantic information of the log.

A Detection Method of Bolts on Axlebox Cover Based on Cascade Deep Convolutional Neural Network

This paper proposes a cascade deep convolutional neural network to address the loosening detection problem of bolts on axlebox covers.Firstly,an SSD network based on ResNet50 and CBAM module by improving bolt image features is proposed for locating bolts on axlebox covers.And then,theA2-PFN is proposed according to the slender features of the marker lines for extracting more accurate marker lines regions of the bolts.Finally,a rectangular approximationmethod is proposed to regularize themarker line regions asaway tocalculate the angle of themarker line and plot all the angle values into an angle table,according to which the criteria of the angle table can determine whether the bolt with the marker line is in danger of loosening.Meanwhile,our improved algorithm is compared with the pre-improved algorithmin the object localization stage.The results show that our proposed method has a significant improvement in both detection accuracy and detection speed,where ourmAP(IoU=0.75)reaches 0.77 and fps reaches 16.6.And in the saliency detection stage,after qualitative comparison and quantitative comparison,our method significantly outperforms other state-of-the-art methods,where our MAE reaches 0.092,F-measure reaches 0.948 and AUC reaches 0.943.Ultimately,according to the angle table,out of 676 bolt samples,a total of 60 bolts are loose,69 bolts are at risk of loosening,and 547 bolts are tightened.

Rapid Detection of Accelerants in Fire Debris Using a Field Portable Mid-Infrared Quantum Cascade Laser Based Analyzer

Arson presents a challenging crime scene for fire investigators worldwide. Key to the investigation of suspected arson cases is the analysis of fire debris for the presence of accelerants or ignitable liquids. This study has investigated the application and method development of vapor phase mid-Infrared (mid-IR) spectroscopy using a field portable quantum cascade laser (QCL) based system for the detection and identification of accelerant residues such as gasoline, diesel, and ethanol in fire debris. A searchable spectral library of various ignitable fluids and fuel components measured in the vapor phase was constructed that allowed for real-time identification of accelerants present in samples using software developed in-house. Measurement of vapors collected from paper material that had been doused with an accelerant followed by controlled burning and then extinguished with water showed that positive identification could be achieved for gasoline, diesel, and ethanol. This vapor phase mid-IR QCL method is rapid, easy to use, and has the sensitivity and discrimination capability that make it well suited for non-destructive crime scene sample analysis. Sampling and measurement can be performed in minutes with this 7.5 kg instrument. This vibrational spectroscopic method required no time-consuming sample pretreatment or complicated solvent extraction procedure. The results of this initial feasibility study demonstrate that this portable fire debris analyzer would greatly benefit arson investigators performing analysis on-site.

Improved pedestrian detection with peer AdaBoost cascade

Focusing on data imbalance and intraclass variation,an improved pedestrian detection with a cascade of complex peer AdaBoost classifiers is proposed.The series of the AdaBoost classifiers are learned greedily,along with negative example mining.The complexity of classifiers in the cascade is not limited,so more negative examples are used for training.Furthermore,the cascade becomes an ensemble of strong peer classifiers,which treats intraclass variation.To locally train the AdaBoost classifiers with a high detection rate,a refining strategy is used to discard the hardest negative training examples rather than decreasing their thresholds.Using the aggregate channel feature(ACF),the method achieves miss rates of 35%and 14%on the Caltech pedestrian benchmark and Inria pedestrian dataset,respectively,which are lower than that of increasingly complex AdaBoost classifiers,i.e.,44%and 17%,respectively.Using deep features extracted by the region proposal network(RPN),the method achieves a miss rate of 10.06%on the Caltech pedestrian benchmark,which is also lower than 10.53%from the increasingly complex cascade.This study shows that the proposed method can use more negative examples to train the pedestrian detector.It outperforms the existing cascade of increasingly complex classifiers.

An attention-based cascade R-CNN model for sternum fracture detection in X-ray images

Fracture is one of the most common and unexpected traumas.If not treated in time,it may cause serious consequences such as joint stiffness,traumatic arthritis,and nerve injury.Using computer vision technology to detect fractures can reduce the workload and misdiagnosis of fractures and also improve the fracture detection speed.However,there are still some problems in sternum fracture detection,such as the low detection rate of small and occult fractures.In this work,the authors have constructed a dataset with 1227 labelled X-ray images for sternum fracture detection.The authors designed a fully automatic fracture detection model based on a deep convolution neural network(CNN).The authors used cascade R-CNN,attention mechanism,and atrous convolution to optimise the detection of small fractures in a large X-ray image with big local variations.The authors compared the detection results of YOLOv5 model,cascade R-CNN and other state-of-the-art models.The authors found that the convolution neural network based on cascade and attention mechanism models has a better detection effect and arrives at an mAP of 0.71,which is much better than using the YOLOv5 model(mAP=0.44)and cascade R-CNN(mAP=0.55).

Automatic character detection and segmentation in natural scene images

We present a robust connected-component (CC) based method for automatic detection and segmentation of text in real-scene images. This technique can be applied in robot vision, sign recognition, meeting processing and video indexing. First, a Non-Linear Niblack method (NLNiblack) is proposed to decompose the image into candidate CCs. Then, all these CCs are fed into a cascade of classifiers trained by Adaboost algorithm. Each classifier in the cascade responds to one feature of the CC. Proposed here are 12 novel features which are insensitive to noise, scale, text orientation and text language. The classifier cascade allows non-text CCs of the image to be rapidly discarded while more computation is spent on promising text-like CCs. The CCs passing through the cascade are considered as text components and are used to form the segmentation result. A prototype system was built, with experimental results proving the effectiveness and efficiency of the proposed method.

Multiscale edge detection of noisy images using wavelets

The classical edge detectors work fine with the high quality pictures, but often are not good enough for noisy images because they cannot distinguish edges of different significance. The paper presented a novel approach to multiscale edge detection for noisy images using wavelet transforms based on Lipschitz regularity coefficients and a cascade algorithm. The relationship between wavelet transform and Lipschitz regularity was established. The proposed wavelet based edge detection algorithm combined the coefficients of wavelet transforms along with a cascade algorithm which significantly improves the result. The comparison between the proposed method and the classical edge detectors was carried out. The algorithm was applied to various images and its performance was discussed. The results of edge detection of contaminated images using the proposed algorithm show that it works better than the classical edge detectors.

Defect Detection Algorithm of Patterned Fabrics Based on Convolutional Neural Network

The background pattern of patterned fabrics is complex,which has a great interference in the extraction of defect features.Traditional machine vision algorithms rely on artificially designed features,which are greatly affected by background patterns and are difficult to effectively extract flaw features.Therefore,a convolutional neural network(CNN)with automatic feature extraction is proposed.On the basis of the two-stage detection model Faster R-CNN,Resnet-50 is used as the backbone network,and the problem of flaws with extreme aspect ratio is solved by improving the initialization algorithm of the prior frame aspect ratio,and the improved multi-scale model is designed to improve detection of small defects.The cascade R-CNN is introduced to improve the accuracy of defect detection,and the online hard example mining(OHEM)algorithm is used to strengthen the learning of hard samples to reduce the interference of complex backgrounds on the defect detection of patterned fabrics,and construct the focal loss as a loss function to reduce the impact of sample imbalance.In order to verify the effectiveness of the improved algorithm,a defect detection comparison experiment was set up.The experimental results show that the accuracy of the defect detection algorithm of patterned fabrics in this paper can reach 95.7%,and it can accurately locate the defect location and meet the actual needs of the factory.

Real-Time Face Detection and Recognition in Complex Background

This paper provides efficient and robust algorithms for real-time face detection and recognition in complex backgrounds. The algorithms are implemented using a series of signal processing methods including Ada Boost, cascade classifier, Local Binary Pattern (LBP), Haar-like feature, facial image pre-processing and Principal Component Analysis (PCA). The Ada Boost algorithm is implemented in a cascade classifier to train the face and eye detectors with robust detection accuracy. The LBP descriptor is utilized to extract facial features for fast face detection. The eye detection algorithm reduces the false face detection rate. The detected facial image is then processed to correct the orientation and increase the contrast, therefore, maintains high facial recognition accuracy. Finally, the PCA algorithm is used to recognize faces efficiently. Large databases with faces and non-faces images are used to train and validate face detection and facial recognition algorithms. The algorithms achieve an overall true-positive rate of 98.8% for face detection and 99.2% for correct facial recognition.

基于改进Cascade R-CNN的安全帽检测算法

针对安全帽检测中,目标形状、尺度变化大,易出现漏检、误检等问题,提出了一种基于改进级联基于区域的卷积神经网络(Cascade R-CNN)的安全帽检测算法。首先,对ResNet50进行改进形成D-ResNet50,利用可变形卷积仅增加少量参数就可增大感受野的特性,对特征提取网络的C2~C5卷积层进行重塑,提高网络对目标几何变换的适应能力和特征提取能力。其次,将D-ResNet50作为主干网络引入Cascade R-CNN,形成级联目标检测器,在每个阶段对正负样本重采样,抑制误检问题。再次,对递归特征金字塔进行改进,更高效地进行多尺度特征融合,并且基于反馈信息对特征进行二次处理,增强特征表达,提高网络的分类和定位能力。最后,使用Soft-非极大值抑制(Soft-NMS)进行后处理,进一步解决漏检问题。提出的方法在Hard hat workers数据集上的AP值相比检测基线提高了3.5%,与Sparse R-CNN、TridentNet、VFnet等先进算法相比分别提升了4.7%、5.9%、2.3%等。

基于改进Cascade R-CNN算法的船舶目标检测方法

为了解决实际场景下船舶目标检测精度低的问题,本文在Cascade R-CNN算法的基础上,提出一种船舶目标检测方法 Boat R-CNN。Boat R-CNN使用带自注意力机制的Swin-Transformer Tiny网络提取图像特征,使用Soft-NMS非极大值抑制方法提升候选框过滤精度,使用Smooth_L1损失函数加速模型收敛并减少梯度爆炸情况,使用CIOU边界框回归损失提高候选框回归质量,并针对船舶目标的形状特征优化锚框的长宽比,提高锚框的生成质量。实验结果表明,Boat R-CNN算法的精度相比原版Cascade R-CNN算法提高了21.8%,相比主流Faster R-CNN算法提高了30.3%,有效提升了实际场景下的船舶目标检测精度。

Design of 1.33 μm and 1.55 μm Wavelengths Quantum Cascade Photodetector

In this paper, a quantum cascade photodetector based on intersubband transitions in quantum wells with ability of detecting 1.33 μm and 1.55 μm wavelengths in two individual current paths is introduced. Multi quantum wells structures based on III-Nitride materials due to their large band gaps are used. In order to calculate the photodetector parameters, wave functions and energy levels are obtained by solving 1-D Schrodinger–Poisson equation self consistently at 80 ?K. Responsivity values are about 22 mA/W and 18.75 mA/W for detecting of 1.33 μm and 1.55 μm wavelengths, respectively. Detectivity values are calculated as 1.17 × 107 (Jones) and 2.41 × 107 (Jones) at wavelengths of 1.33 μm and 1.55 μm wavelengths, respectively.

基于Cascade R-CNN的乳腺钼靶肿块检测算法研究

乳腺癌生物学特性复杂,恶性程度极高,位于女性恶性肿瘤发病率首位。乳腺钼靶肿块的X射线检查是早期确诊乳腺癌的重要方式。但乳腺钼靶肿块的检测尚处于早期阶段,现有的计算机辅助检测检测精度较低。针对这一问题,论文提出了一种基于Cascade R-CNN的乳腺钼靶肿块检测方法。实验使用南佛罗里达大学的乳房X光检查数据集,将乳腺钼靶肿块分为良性和恶性两类。通过在特征网络中加入注意力模块,提取了较为丰富的乳腺钼靶肿块特征。此外,论文提出了一种新的FPN网络FA-FPN,进一步提高了乳腺钼靶肿块病灶特征的提取,解决了深层网络在下采样中特征出现稀释的问题,提高了乳腺钼靶肿块的检测准确率。经实验验证,该模型在南佛罗里达大学的乳房X光检查数据集上的mAP值达到82.9%,在AP75下表现尤为突出。该方法在乳腺钼靶肿块的检测中具有良好的性能,可以提高乳腺钼靶肿块的检测精度,并在一定程度上避免了误检和漏检。

改进的Cascade R-CNN算法在目标检测上的应用

针对Cascade R-CNN目标检测算法中存在检测精度较低以及目标遮挡问题,本文提出一种改进的Cas-cade R-CNN网络目标检测算法.该算法在主干网络ResNet101中引入可切换空洞卷积模块(Switchable Atrous Convolu-tion,SAC),该模块主要由两个全局上下文模块以及SAC组件构成,采用SAC组件以不同的空洞卷积率对特征进行卷积,并使用Switch函数收集特征来提高特征提取能力.同时,在ResNet101残差网络中引入坐标注意力机制(Coordi-nate Attention,CA),该机制将位置信息嵌入通道注意力中,用于更好地获取方向感知和位置感知信息,进而提高目标检测精度.此外,针对目标遮挡问题,引入Repulsion Loss损失函数.该损失函数主要由吸引项和排斥项组成,吸引项使得预测框和匹配上的目标框尽可能接近,排斥项使得预测框远离错误目标,进而减少非极大值抑制(Non-Maximum Suppression,NMS)的误检,提高目标检测中遮挡问题的检测精度.实验结果表明,在公开的科大讯飞AI挑战赛数据集上,与原算法测试性能相比,改进的Cascade R-CNN网络对该数据集检出率增长了2.39%,改进算法的识别精度有一定的提高.

A learning-based method to detect and segment text from scene images

This paper proposes a learning-based method for text detection and text segmentation in natural scene images. First, the input image is decomposed into multiple connected-components (CCs) by Niblack clustering algorithm. Then all the CCs including text CCs and non-text CCs are verified on their text features by a 2-stage classification module, where most non-text CCs are discarded by an attentional cascade classifier and remaining CCs are further verified by an SVM. All the accepted CCs are output to result in text only binary image. Experiments with many images in different scenes showed satisfactory performance of our proposed method.

基于改进Cascade R-CNN的绝缘子故障检测方法研究

针对航拍图中存在的绝缘子故障位置在图像中占比小、背景环境复杂导致的故障检测准确率低的问题,提出一种基于改进Cascade R-CNN模型的绝缘子故障检测方法。在原有Cascade R-CNN模型的基础上,在骨干网络中引入可变形卷积学习几何变换能力,在检测器中引入平衡损失函数平衡难易样本。在模型训练阶段,使用Copy-Paste与Mosica丰富故障绝缘子样本,平衡正负样本。使用该模型对航拍绝缘子图片进行故障检测实验,改进损失函数的模型与传统Cascade R-CNN模型相比平均召回率提升0.38%,引入可变卷积后的Cascade R-CNN模型与Faster R-CNN模型,相比平均召回率,从原来的89.78%变成93.49%,结果表明该模型能够有效克服样本遮挡以及样本不平衡的干扰。

Skin Detection Method Based on Cascaded AdaBoost Classifier

Skin detection has been considered as the principal step in many machine vision systems,such as face detection and adult image filtering.Among all these techniques,skin color is the most welcome cue because of its robustness.However,traditional color-based approaches poorly perform on the classification of skin-like pixels.In this paper,we propose a new skin detection method based on the cascaded adaptive boosting(AdaBoost) classifier,which consists of minimum-risk based Bayesian classifier and models in different color spaces such as HSV(hue-saturation-value),YCgCb(brightness-green-blue) and YCgCr(brightness-green-red).In addition,we have constructed our own database that is larger and more suitable for training and testing on filtering adult images than the Compaq data set.Experimental results show that our method behaves better than the state-ofthe-art pixel-based skin detection techniques on processing images with skin-like background.

基于改进Cascade R-CNN的布匹瑕疵检测算法

布匹瑕疵的自动化检测是目前纺织行业面临的一个难点问题。针对当前布匹瑕疵检测算法对尺度和长宽比变化大、小目标众多的样本检测效果并不理想的问题,提出了基于改进Cascade R-CNN网络的布匹瑕疵检测算法。首先,在特征提取网络ResNet-50中融入可变形卷积,自适应地提取更多的瑕疵形状与尺度特征;其次,在特征金字塔网络上采样前引入平衡特征金字塔,缩小特征融合前各特征层之间的语义差距,得到更具表达力的多尺度特征;然后,根据瑕疵尺度与长宽比特点重新设计更适合的初始锚框;最后,采用具有尺度不变性的GIoU Loss作为级联检测器的回归损失,以获取更加精确的瑕疵预测边界框。实验结果表明,相比基于Cascade R-CNN的算法,改进后的Cascade R-CNN算法对布匹瑕疵检测的平均精确率获得了明显提升。

基于改进Cascade R-CNN的探地雷达管线目标检测

针对人工识别探地雷达管线图像时效率低、误差大和成本高昂等问题,本文提出了一种基于改进Cascade R-CNN的管线目标智能化检测方法.首先对探地雷达管线图像数据集进行预处理,提升数据质量.然后采用ResNeXt代替ResNet作为主干网络提取目标特征信息,并添加多尺度特征融合模块FPN使高层特征向低层特征融合,增强低层特征表达能力.其次,使用高斯形式的非极大值抑制方法Soft-NMS得到更加精准的候选框,使用Smooth_L1作为损失函数,加速了模型收敛并且降低了训练中发生梯度爆炸的概率.最后,对于管线目标特殊的形状特征,设置合适的锚框长宽比和大小,提高锚框的生成质量.实验结果表明,本文方法解决了复杂特征的地下管线目标智能化检测,对地下管线目标检测的平均精度达到94.7%,比Cascade R-CNN方法提高了10.1%.

基于改进Cascade R-CNN模型的机器人抓取检测研究

为提高多物体抓取检测网络的抓取检测准确率,提出一种基于改进Cascade R-CNN模型的机械臂抓取检测算法.首先,引入ResNeXt结构能够在不加大网络设计难度的前提下提高了模型的准确率;引入带空洞卷积的空间金字塔池化模块以解决分辨率较低的问题;接着对抓取框回归分支和角度分类分支以分治方法进行优化.其次,针对多物体抓取数据集缺乏的问题,构建多目标抓取数据集(multi-object grasping dataset,MOGD),有效地扩充了多物体抓取检测数据集.最后,基于改进Cascade R-CNN模型设计抓取检测网络,实验结果表明,改进后的算法效率更高,PI-Cascade R-CNN实验准确率为93%,较Cascade R-CNN提升1.5个百分点.