Improved Weighted Local Contrast Method for Infrared Small Target Detection

In order to address the problem of high false alarm rate and low probabilities of infrared small target detection in complex low-altitude background,an infrared small target detection method based on improved weighted local contrast is proposed in this paper.First,the ratio information between the target and local background is utilized as an enhancement factor.The local contrast is calculated by incorporating the heterogeneity between the target and local background.Then,a local product weighted method is designed based on the spatial dissimilarity between target and background to further enhance target while suppressing background.Finally,the location of target is obtained by adaptive threshold segmentation.As experimental results demonstrate,the method shows superior performance in several evaluation metrics compared with six existing algorithms on different datasets containing targets such as unmanned aerial vehicles(UAV).

Renal blood perfusion in GK rats using targeted contrast enhanced ultrasonography

Objective: To explore application of targeted contrast enhanced ultrasonography in diagnosis of early stage vascular endothelial injury and diabetic nephropathy. Methods: Targeted Sono VueTM microbubble was prepared by attaching anti-TM monoclonal antibody to the surface of ordinary micro-bubble Sono Vue by biotin-avidin bridge method and ultrasonic instrument was used to evaluate the developing situation of targeted microbubble in vitro. Twenty 12-week-old male GK rats and 20 Wistar rats were enrolled in this study, and were randomly divided into targeted angiography group and ordinary angiography group. Targeted microbubbles Sono VueTM or general microbubble Sono Vue were rapidly injected to the rats via tail vein; the developing situation of the two contrast agents in rats kidneys was dynamically observed. Time intensity curve was used to analyse rat kidney perfusion characteristics in different groups. Results: Targeted ultrasound microbubble Sono Vue-TM was successfully constructed, and it could be used to develop an external image. Targeted microbubbles Sono Vue-TM enabled clear development of experimental rat kidney. Time intensity curve shapes of rat kidney of the two groups showed as single apex with steep ascending and slowly descending branch. Compared with the control group, the rising slope of the GK rat renal cortex, medulla in targeted angiography group increased(P<0.05); the peak intensity of medulla increased(P<0.05), and the total area under the curve of medulla increased(P<0.05). Compared with control group, the ascending branch of the GK rat in renal cortex, medulla in ordinary angiography group increased(P<0.05). The peak intensity of the curve increased(P<0.05), and the total area under the curve increased(P<0.05). Compared with the ordinary angiography group, the peak of GK rat medullacurve in targeted angiography group intensity increased(P<0.05), and the total area under the curve increased(P<0.05). Conclusions: Targeted microbubbles Sono VueTM can make a clear development of experimental rat kidney, its stable performance meet the requirement of ultrasonic observation time limit, and it can reflect early changes of blood perfusion in GK rat kindey.

Contrast-enhanced and targeted ultrasound

Ultrasonic imaging is becoming the most popular medical imaging modality,owing to the low price per examination and its safety.However,blood is a poor scatterer of ultrasound waves at clinical diagnostic transmit frequencies.For perfusion imaging,markers have been designed to enhance the contrast in B-mode imaging.These so-called ultrasound contrast agents consist of microscopically small gas bubbles encapsulated in biodegradable shells.In this review,the physical principles of ultrasound contrast agent microbubble behavior and their adjustment for drug delivery including sonoporation are described.Furthermore,an outline of clinical imaging applications of contrast-enhanced ultrasound is given.It is a challenging task to quantify and predict which bubble phenomenon occurs under which acoustic condition,and how these phenomena may be utilized in ultrasonic imaging.Aided by high-speed photography,our improved understanding of encapsulated microbubble behavior will lead to more sophisticated detection and delivery techniques.More sophisticated methods use quantitative approaches to measure the amount and the time course of bolus or reperfusion curves,and have shown great promise in revealing effective tumor responses to anti-angiogenic drugs in humans before tumor shrinkage occurs.These are beginning to be accepted into clinical practice.In the long term,targeted microbubbles for molecular imaging and eventually for directed anti-tumor therapy are expected to be tested.

Moving target detection based on improved ghost suppression and adaptive visual background extraction

Visual background extraction algorithm(ViBe)uses the first frame image to initialize the background model,which can easily introduce the“ghost”.Because ViBe uses the fixed segmentation threshold to achieve the foreground and background segmentation,the detection results in many false detections for the highly dynamic background.To solve these problems,an improved ghost suppression and adaptive Visual Background Extraction algorithm is proposed in this paper.Firstly,with the pixel’s temporal and spatial information,the historical pixels of a certain combination are used to initialize the background model in the odd frames of the video sequence.Secondly,the background sample set combined with the neighborhood pixels are used to determine a complex degree of the background,to acquire the adaptive segmentation threshold.Thirdly,the update rate is adjusted based on the complexity of the background.Finally,the detected result goes through a post-processing to achieve better detection results.The experimental results show that the improved algorithm will not only quickly suppress the“ghost”,but also have a better detection in a complex dynamic background.

Studies on Hepatocyte-Targeting Magnetic Resonance Imaging Macromolecular Contrast Media

DTPA was covalently conjugated to the copolymer of L lysine and L tyrosine, then transchelated with Gd EDTA,the resultant polymer chelates were further reacted with 6 O bromoacetyl D galactose or methyl lactobionyloxyacetate to give PLT(Gd DTPA) with galactose moiety as liver targeting group.These macromolecular MRI contrast agents were characterized by means of FT IR, ICP AES, elementary analyses and UV Vis spectrophotometry. Their in vitro relaxivity and liver targeting property in mice were also investigated.

ACCURATE DETECTION OF HIGH-SPEED MULTI-TARGET VIDEO SEQUENCES MOTION REGIONS BASED ON RECONSTRUCTED BACKGROUND DIFFERENCE

The paper first discusses shortcomings of classical adjacent-frame difference. Sec ondly, based on the image energy and high order statistic(HOS) theory, background reconstruction constraints are setup. Under the help of block-processing technology, background is reconstructed quickly. Finally, background difference is used to detect motion regions instead of adjacent frame difference. The DSP based platform tests indicate the background can be recovered losslessly in about one second, and moving regions are not influenced by moving target speeds. The algorithm has important usage both in theory and applications.

Application of ultrasound microbubble contrast technology in ophthalmic targeted therapy:literature analysis

AIM: To analyze the application of microbubble contrast technology in the treatment of ophthalmic diseases, mainly analyzing its advantages and existing problems. METHODS: A total of 30 representative literatures about the application of ultrasound contrast agent in gene targeted therapy at home and abroad were collected, and focusing on sorting out the literature reporting the treatment of ophthalmic diseases with microbubble contrast technology in recent years, then recalling its advantages and problems, finally making reasonable assessment on existing problems and proposing possible solutions to the problems. RESULTS: Due to its unique safety and efficacy, the treatment of ophthalmic diseases with microbubble contrast technology has increasingly drawn the attention of clinicians, but two relevant issues should be considered: first, the nature of contrast agent and the choice of corresponding ultrasound parameters; second, relative incidence of tissue bleeding, intravascular hemolysis, moderate or severe allergy as well as other side effects. CONCLUSION: Microbubble may become the carrier of targeted therapy, and as a kind of new non-invasive delivery system, the ultrasound contrast agent has broad application prospects, but its application in ophthalmic research is still in its initial stage and the safety of contrast-enhanced ultrasound still needs further study.

A pixel-level local contrast measure for infrared small target detection

Infrared(IR) small target detection is one of the key technologies of infrared search and track(IRST)systems. Existing methods have some limitations in detection performance, especially when the target size is irregular or the background is complex. In this paper, we propose a pixel-level local contrast measure(PLLCM), which can subdivide small targets and backgrounds at pixel level simultaneously.With pixel-level segmentation, the difference between the target and the background becomes more obvious, which helps to improve the detection performance. First, we design a multiscale sliding window to quickly extract candidate target pixels. Then, a local window based on random walker(RW) is designed for pixel-level target segmentation. After that, PLLCM incorporating probability weights and scale constraints is proposed to accurately measure local contrast and suppress various types of background interference. Finally, an adaptive threshold operation is applied to separate the target from the PLLCM enhanced map. Experimental results show that the proposed method has a higher detection rate and a lower false alarm rate than the baseline algorithms, while achieving a high speed.

A camouflage target detection method based on local minimum difference constraints

To address the problems of missing inside and incomplete edge contours in camouflaged target detection results,we propose a camouflaged moving target detection algorithm based on local minimum difference constraints(LMDC).The algorithm first uses the mean to optimize the initial background model,removes the stable background region by global comparison,and extracts the edge point set in the potential target region so that each boundary point(seed)grows along the center of the target.Finally,we define the minor difference constraints term,combine the seed path and the target space consistency,and calculate the attributes of each pixel in the potential target area to realize camouflaged moving target detection.The algorithm of this paper is verified based on a public data sofa video and test videos and compared with the five classic algorithms.The experimental results show that the proposed algorithm yields good results based on integrity,accuracy,and a number of objective evaluation indexes,and its overall performance is better than that of the compared algorithms.

复杂背景下基于YOLOv7-tiny的图像目标检测算法

“黑飞”无人机一旦带有炸弹等物品,会对人们带来威胁。对在公园、游乐场、学校等复杂背景下“黑飞”的无人机进行目标检测是十分必要的。前沿算法YOLOv7-tiny属于轻量级网络,具有更小的网络结构和参数,更适合检测小目标,但在识别小目标无人机时出现特征提取能力弱、回归损失大、检测精度低的问题;针对此问题,提出了一种基于YOLOv7-tiny改进的无人机图像目标检测算法YOLOv7-drone。首先,建立无人机图像数据集;其次,设计一种新的注意力机制模块SMSE嵌入到特征提取网络中,增强对复杂背景下无人机目标的关注度;然后,在主干网络中融入RFB结构,扩大特征层的感受野,丰富特征信息以增强特征提取的鲁棒性;然后,改进网络中的特征融合机制,通过新增小目标检测层,增加对小尺度目标的检测精度;然后,改变损失函数提高模型的收敛速度,减少损失以增强模型的鲁棒性;最后,引入可变形卷积(Deformable convolution, DCN),更好的根据目标本身形状进行特征提取,提升了检测精度。在PASCAL VOC公共数据集上进行对比实验,结果表明改进后的算法YOLO7-drone相比于YOLOv7-tiny,平均精度(map@0.5)提升了6%;在自制无人机数据集上进行实验,结果表明YOLOv7-drone与原算法相比,平均精度(map@0.5)提高了6.1%,并且检测速度为72帧/s;与YOLOv5l、YOLOv7目标检测算法进行对比实验,结果表明改进后的算法在平均精度(map@0.5)上分别高于对比算法4%、3.1%,验证了文中算法的可行性。

稀疏约束与时间一致的背景感知相关滤波目标跟踪

背景感知滤波算法通过循环移位采集真实负样本,有效解决了边界效应.但在复杂场景例如遮挡、快速移动、背景干扰等,其较大的采样区域导致过多背景在杂波干扰,从而影响跟踪效果.针对这一问题,本文首先提取灰度HOG特征与颜色CN特征来提高目标外观模型,在基准目标函数基础上引入L1稀疏正则约束形成弹性网络以自适应筛选关键特征,增强滤波器在复杂背景下的判别能力.同时针对BACF在跟踪过程中目标快速变化,本文引入时间正则项提高滤波器抑制畸变的能力.最后,本文提出了一种独立的尺度滤波器算法,准确提供目标尺度大小.实验仿真结果表明,在公开数据集OTB-2013和OTB-2015上,本文算法较基准算法有很大提升,能够较好应对不同复杂场景下的跟踪难题.

基于联合方向梯度和均值对比度的红外弱小目标检测方法

针对小目标在整幅图像中占比很低,且目标周围存在大量杂波,提出了一种基于联合方向梯度(Associated Directional Gradient,ADG)和均值对比度(Mean Contrast,MC)的红外弱小目标检测算法。该算法由两个模块组成:ADG利用红外弱小目标的高斯分布模型,将单一方向的梯度与一个相邻方向上的梯度相加组成新的联合梯度特征,增强真实目标、抑制背景杂波的同时能够消除高亮边缘对目标检测效果的影响;MC融入方向信息来计算目标的多方向对比度,选用多方向对比度的最小值抑制结构噪声,并将均值滤波的思想引入对比度的计算,抑制背景中的孤立噪声,进一步降低检测的虚警率。在实际红外图像上的实验结果表明,该算法在增强目标信噪比和抑制背景噪声方面,能够取得较好效果。

一种基于目标与背景特征分离模型的高光谱目标检测修正算法

高光谱图像立方体数据可以提供成像场景中地物在可见光和近红外波长范围内的空间信息和地物属性诊断的光谱特征信息,在目标检测与识别方面拥有得天独厚的天然优势。然而,基于高光谱图像数据的目标检测也存在一定缺陷,如经典的高光谱目标检测算法仅利用光谱维度信息检测目标,检测模型要么对背景高维特征矩阵构建的准确度不足,要么对背景先验光谱特征的完备性要求较高,导致算法对不同复杂度的检测场景适应性不强。因此,基于计算复杂度较低、参数需求量较少且检测性能较为优异的经典多目标检测算法—多目标约束能量最小化(MCEM),提出了一种基于目标与背景环境特征分离模型的高光谱目标检测修正算法(R-MCEM)。首先,设计了一个与目标形状、尺寸相近的逐像元移动运算窗口,依次计算窗口中的每个像元与窗口内其他像元的光谱距离之和D1,像元与各类目标的光谱距离之和D2。其次,采用获得D1/D2最小值的像元替换窗口内的所有像元值。然后,自左向右、自上而下逐像元移动窗口,重复窗口内每一个像元与目标、背景像元的光谱距离运算,并确定窗口内与背景相似度最高、与目标相似度最低的像元。直到移动运算窗口遍历整个高光谱图像,大幅提升了基于目标与背景环境特征分离的背景高维特征矩阵准确度。分别设计了基于实测高光谱图像数据和模拟图像数据的修正检测算法性能验证试验,并采用三维操作特征曲线(3D ROC)结合目标与背景分离度(SDBT)开展修正算法的检测精度评估。试验结果表明,提出的修正算法有效减少了虚警率,提高了检测精度。基于实测数据的检测精度、目标与背景分离度由MCEM算法的0.937 7、 0.57提升到R-MCEM的0.993 5、 0.67,基于模拟数据的亚像元检测能力由MCEM的20%丰度提升到R-MCEM的15%丰度。

气体微泡光声造影剂的靶向修饰与应用

气体微泡作为一种新型的光声造影剂,由于其具有生物相容性良好、成像灵敏度高、易于修饰等优点,在生物医学成像领域引起了广泛的关注。然而,气体微泡的非靶向性限制了其在临床上的应用。为了提高气体微泡的光声造影剂的靶向性和特异性,需要对其进行靶向修饰。本文综述气体微泡的光声造影剂的靶向修饰方法,包括物理修饰法、化学修饰法和生物修饰法,阐述气体微泡的光声造影剂的靶向修饰在肿瘤成像、炎症成像、感染成像和血管成像中的应用,分析气体微泡的光声造影剂的靶向修饰的优势和局限性,展望未来的发展方向。

利用局部特征匹配的运动小目标光流估计

基于深度光流估计的动态背景运动小目标检测,为了保证小目标的检测性能,一般采用较少的下采样次数以维持较高的分辨率,但由此带来了较大的计算耗时。特征匹配是深度光流估计的一个核心处理环节,其耗时在光流估计整体耗时中的占比较大,且对下采样次数非常敏感。据此,提出一种基于局部特征匹配的快速光流估计算法:引入目标运动信息,缩小特征匹配的空间范围,减少待处理的数据量;设计分块局部匹配策略,引入批处理机制,避免出现逐点局部匹配策略数据处理耗时过大问题,实现算法加速。在此基础上,在光流估计获取的光流场上,采用CenterNet网络检测运动目标对应的光流异常区域。从光流估计耗时、检测精度等方面开展了实验验证,结果表明:针对运动小目标检测,分块特征匹配光流估计比全局特征匹配光流估计耗时减少约25%,目标检测性能相当。

DCE-MRI与超声和钼靶X线在乳腺癌诊断价值对比

目的:对比动态增强磁共振成像(DCE-MRI)与超声、钼靶X线对乳腺癌的诊断价值。方法:选取2021年3月~2024年3月疑似乳腺癌患者80例,均进行DCE-MRI、超声、钼靶X线检测,以病理学结果为金标准,比较三种方法的诊断效能。结果:依据金标准,DCE-MRI灵敏度93.22%、特异度76.19%、准确度88.75%、阳性预测值91.67%、阴性预测值80.00%均优于超声、钼靶X线(P<0.05)。结论:DCE-MRI在乳腺癌诊断中具有较高的诊断效能,值得在临床上推广。

基于改进顶帽变换的红外弱小目标检测

天空背景下的红外弱小目标检测技术较为成熟,但在近地复杂背景下,红外弱小目标的检测存在准确率不高、虚警目标多、实时性差的问题。针对以上问题,该文提出一种基于改进顶帽变换的红外弱小目标检测算法(OTHOLCM)。该算法采用基于改进顶帽变换的图像预处理算法(OTH),通过对不同灰度值的图像采取不同的策略针对性地处理图像,达到目标增强、背景抑制的效果。并在此基础上,采用基于改进多尺度局部对比度的红外弱小目标检测算法(OLCM),通过针对目标尺寸特点进行尺度设计,使得在保证算法实时性的基础上扩大目标尺寸检测范围。实验证明:OTHOLCM算法可以保证实时性并明显提高目标检测准确率、减少虚警目标数量。与3层模板局部差异度量算法(TTLDM)、基于边角感知的时空张量模型(ECASTT)等先进算法相比,OTHOLCM算法可使真阳性率分别提高近79%,61%,假阳性率分别降低近77%,73%,目标检测速度达到每秒25帧。

改进YOLOv5s的弱光水下生物目标检测算法

针对水下光学图像目标检测过程中由于水中光线衰弱严重、图像环境复杂和拍摄设备移动等造成的生物识别精度低的问题,提出了基于改进YOLOv5s的弱光水下生物目标实时检测算法YOLOv5s-underwater。针对弱光水下光线衰弱的问题,引入了限制对比度自适应直方图均衡(CLAHE)算法对输入图像进行预处理,解决了颜色失真和图像毛糙的问题。针对复杂的弱光水下图像环境,提出了快速空间金字塔池化(SPPF)模块,解决了水下物体区分度低和特征损失严重的问题。针对拍摄设备移动带来的场景和形态变化问题,提出了一种基于旋转窗口的Swin-Transformer模块,提高了模型的泛化能力。针对水下小目标,修改了网络模型结构,提高了小目标的检测能力。仿真和实验结果表明:所提算法相较于YOLOv5s检测精度提高30.7%,证明了算法的有效性。

基于结构相似度和鲁棒主成分分析的运动目标检测

运动目标传统检测方法只考虑图像的亮度或纹理等某一种特性,受特异值影响较大,对噪声比较敏感,鲁棒性也不够好,而且背景恢复精度不高。针对以上局限性,提出一种融合结构相似度(structural similarity,SSIM)全参考模型和鲁棒主成分分析(robust principal component analysis,RPCA)的运动目标检测方法。此方法综合考虑图像的亮度、对比度和结构三种特性,不采用传统的背景减除法,而是把图像像素点的结构相似度作为度量来实现运动对象与背景的分离。实验结果表明,此方法准确率可达0.95,且F度量较传统运动目标检测算法平均提升0.15,总体上比传统方法更具优势。

基于融合编码器与双解码器的半监督疲劳裂纹分割

针对现有基于深度学习的疲劳裂纹分割算法依赖于大量像素级标记的问题,提出了一种半监督疲劳裂纹分割网络SFD-Net.SFD-Net利用对比学习方法进行半监督训练,以减少对大量像素级标记的依赖.同时,它采用融合编码器和双解码器的设计,旨在更好地捕捉裂纹区域的特征并提高分割准确性.通过引入改进注意力模块和边界优化模块,增强了对裂纹特征的表示,提高了裂纹边界的分割质量.在公开的疲劳裂纹数据集上对SFD-Net进行性能验证.结果表明:相较于使用相同标记比例的全监督算法,SFD-Net的分割性能有明显提升;仅使用25%的标记数据时,SFD-Net的交并比(IoU)达到70.6%,超过了使用100%标记数据的其他全监督算法的平均IoU(69.1%);同时,与其他先进的半监督方法相比,SFD-Net在所有标记数据比例下均取得了最高的预测精度.