Búsqueda Imágenes Maps Play YouTube Noticias Gmail Drive Más »
Iniciar sesión
Usuarios de lectores de pantalla: deben hacer clic en este enlace para utilizar el modo de accesibilidad. Este modo tiene las mismas funciones esenciales pero funciona mejor con el lector.

Patentes

  1. Búsqueda avanzada de patentes
Número de publicaciónCN103027713 A
Tipo de publicaciónSolicitud
Número de solicitudCN 201210563056
Fecha de publicación10 Abr 2013
Fecha de presentación22 Dic 2012
Fecha de prioridad22 Dic 2012
Número de publicación201210563056.9, CN 103027713 A, CN 103027713A, CN 201210563056, CN-A-103027713, CN103027713 A, CN103027713A, CN201210563056, CN201210563056.9
Inventores芦祎, 李济舟, 周永进, 刘骏识, 王磊
Solicitante中国科学院深圳先进技术研究院
Exportar citaBiBTeX, EndNote, RefMan
Enlaces externos:  SIPO, Espacenet
Muscle thickness measuring method and system based on ultrasonic image
CN 103027713 A
Resumen
The invention discloses a muscle thickness measuring method and system based on an ultrasonic image. The method comprises the following steps of: extracting an interesting image from the ultrasonic image; acquiring the positions of a plurality of initial tracing windows selected in the interesting image; tracing the tracing windows, and determining the positions of a plurality of tracing windows which correspond to every frame of subsequent image through a tracing algorithm; processing tracing windows which are similar to the modes of surrounding images in the tracing windows of each frame of image by taking a diagonal intersection point as a central point, and processing other tracing windows in every frame of image by adopting an edge detection method; and computing a maximum vertical distance between the position of each tracing window processed by using the central point in every frame of image and the position of each tracing window processed through the edge detection method, wherein the maximum vertical distance is taken as a muscle thickness value. Due to the adoption of the muscle thickness measuring method and the muscle thickness measuring system based on the ultrasonic image, the measuring accuracy and measuring efficiency are increased, and the aim of measuring in real time is fulfilled.
Reclamaciones(10)  traducido del chino
1. 一种基于超声图像的肌肉厚度测量方法,包括以下步骤: 从超声图像中提取感兴趣图像; 获取在所述感兴趣图像中选择的多个初始跟踪窗口的位置; 对多个跟踪窗口进行跟踪,并通过跟踪算法确定后续每帧图像相应的多个跟踪窗口的位置; 对每帧图像的多个跟踪窗口中与周围图像模态相似的跟踪窗口采用取对角线交点为中心点进行处理,对每帧图像中其余跟踪窗口采用边缘检测法进行处理; 计算每帧图像中经过中心点处理后的跟踪窗口的位置与每个经过边缘检测法处理后的跟踪窗口的位置之间的最大垂直距离,将所述最大垂直距离作为经过中心点处理后的跟踪窗口与经过边缘检测法处理后的跟踪窗口之间的肌肉厚度值。 A muscle thickness measurement method based on ultrasound images, comprising the steps of: extracting an image of interest from the ultrasound image; acquiring a selected position in the image of interest in more than one initial tracking window; a plurality of tracking window tracking and follow-up of each image to determine a corresponding plurality of tracking algorithm by tracking the position of the window; for each frame of the plurality of tracking window surrounding the image modality similar tracking window using diagonal intersection take center for processing maximum vertical position and edge detection process after tracking window between each calculated for each frame image after processing center position after tracking window; for every remaining trace window frames using edge detection method for processing distance, the maximum vertical distance as the track through the center of treatment and after the window has elapsed between the trace window edge detection process after muscle thickness value.
2.根据权利要求1所述的基于超声图像的肌肉厚度测量方法,其特征在于,所述跟踪算法为压缩跟踪算法或互相关跟踪算法。 The ultrasound images based on muscle thickness measurement method according to claim 1, characterized in that said tracking algorithm compression tracking algorithms or correlation tracking algorithm.
3.根据权利要求1所述的基于超声图像的肌肉厚度测量方法,其特征在于,所述跟踪算法为压缩跟踪算法; 所述对多个跟踪窗口进行跟踪,并通过跟踪算法确定后续每帧图像相应的多个跟踪窗口的位置的步骤为: 对跟踪窗口所在帧图像进行采样,得到属于所述跟踪窗口位置范围内的样本集合; 采用稀疏矩阵对样本集合中的每个样本进行降维处理,得到压缩特征向量; 对所述压缩特征向量采用分类器进行分类; 从所述样本集合中进行抽样得到两组图像样本; 对所述两组图像样本提取哈尔特征,并采用所述分类器迭代得到后续相邻帧图像相应的跟踪窗口的位置。 3. The ultrasound images based on muscle thickness measurement method according to claim 1, characterized in that the tracking algorithm to compress tracking algorithm; a plurality of tracking window for tracking and follow-up of each frame is determined by the tracking algorithm Step position corresponding to a plurality of tracking window: tracking window where the frame image is sampled to obtain a sample belonging to the tracking range of positions within the window set; sparse matrix sample collection to reduce the dimension of each sample, is compressed feature vector; the compressed feature vectors, classification for classification; sample from the sample collection to obtain two sets of image samples; extracting features of the two images Hal sample, and using the classification iterations to get the position of the subsequent neighboring frame images corresponding trace window.
4.根据权利要求3所述的基于超声图像的肌肉厚度测量方法,其特征在于,所述对所述压缩特征向量采用分类器进行分类的步骤包括: 对所述压缩特征向量采用朴素贝叶斯分类器分类,且分类器中的条件概率满足高斯正态分布。 According to claim thickness measurement method based on muscle ultrasound images, wherein 3, wherein the classifier using the compression feature vectors step of classifying includes: the compression feature vectors, Naive Bayes classifiers, conditional probability and the classification of a Gaussian normal distribution.
5.根据权利要求1所述的基于超声图像的肌肉厚度测量方法,其特征在于,在所述从捕捉的超声图像中提取感兴趣图像的步骤之后,还包括步骤: 对所述感兴趣图像进行预处理,包括: 对所述感兴趣图像进行灰度变换及调整图像对比度。 According to claim thickness measurement method based on muscle ultrasound images, wherein 1, after the step of extracting from the captured image of interest in the ultrasound image, further comprising the step of: images of interest pretreatment, comprising: gray-scale transformation of the image of interest and adjust image contrast.
6. 一种基于超声图像的肌肉厚度测量系统,其特征在于,包括: 提取模块,用于从捕捉的超声图像中提取感兴趣图像; 获取模块,用于获取在所述感兴趣图像中选择的多个初始的跟踪窗口的位置; 跟踪模块,用于对多个跟踪窗口进行跟踪,并通过跟踪算法确定后续每帧图像相应的多个跟踪窗口的位置; 处理模块,用于对每帧图像的多个跟踪窗口中与周围图像模态相似的跟踪窗口采用取对角线交点为中心点进行处理,对每帧图像中其余跟踪窗口采用边缘检测法进行处理; 计算模块,用于计算每帧图像中经过中心点处理后的跟踪窗口的位置与每个经过边缘检测法处理后的跟踪窗口的位置之间的最大垂直距离,将所述最大垂直距离作为经过中心点处理后的跟踪窗口与经过边缘检测法处理后的跟踪窗口之间的肌肉厚度值。 An ultrasound image based on the thickness of the muscle measuring system, characterized by comprising: extraction module for extracting from the captured image of interest in the ultrasound image; acquiring module, for acquiring the selected image of interest location tracking window of the plurality of initial; tracking module for tracking multiple windows to track and determine the location of each subsequent frame image corresponding to a plurality of tracking window by tracking algorithm; processing module for each frame multiple tracking window surrounding the image modality similar tracking window using diagonal intersection take center for processing, for each frame remaining trace window treatment using edge detection method; calculation module for calculating each frame After treatment in the center position after tracking window and the maximum vertical distance between the position after edge detection processing window between each track, the maximum vertical distance as the track through the center of treatment and after the window past the edge Trace window muscle thickness values detected between treatment.
7.根据权利要求6所述的基于超声图像的肌肉厚度测量系统,其特征在于,所述跟踪算法为压缩跟踪算法或互相关跟踪算法。 According to claim ultrasound images based on muscle thickness measurement system 6, characterized in that said tracking algorithm compression tracking algorithms or correlation tracking algorithm.
8.根据权利要求6所述的基于超声图像的肌肉厚度测量系统,其特征在于,所述跟踪算法为压缩跟踪算法; 所述跟踪模块包括: 采样模块,用于对跟踪窗口所在帧图像进行采样,得到属于跟踪窗口位置范围内的样本集合; 降维模块,用于采用稀疏矩阵对样本集合中的每个样本进行降维处理,得到压缩特征向量; 分类模块,用于对所述压缩特征向量采用分类器进行分类; 抽样模块,用于从所述样本集合中进行抽样得到两组图像样本; 迭代模块,用于对所述两组图像样本提取哈尔特征,并采用分类器迭代得到相邻的后续图像帧的跟踪窗口的位置。 8. The ultrasound images based on muscle thickness measuring system according to claim 6, wherein said tracking algorithm compression tracking algorithm; the tracking module comprises: sampling module for tracking window frames where sampling to obtain samples of the trace window position belongs to the range set; dimension reduction module for sparse matrix sample collection to reduce the dimension of each sample to give compression feature vector; classification module for the compression feature vector the use of classification for classification; sampling module for sampling from the sample set of image samples obtained in the two groups; iteration module for extracting features of the two images Hal samples and obtained using an iterative classification Trace window location adjacent subsequent image frames.
9.根据权利要求8所述的基于超声图像的肌肉厚度测量系统,其特征在于,所述分类模块还用于对所述压缩特征向量采用朴素贝叶斯分类器分类,且分类器中的条件概率满足高斯正态分布。 Based muscle according to claim 8, wherein the thickness measurement system ultrasound images, wherein the classification module is also used for the compression feature vectors, naive Bayesian classifier and classifier conditions the probability of a Gaussian normal distribution.
10.根据权利要求6所述的基于超声图像的肌肉厚度测量系统,其特征在于,所述系统还包括: 预处理模块,用于对所述感兴趣图像进行预处理,所述预处理包括对所述感兴趣图像进行灰度变换及调整图像对比度。 10. The ultrasound image based on the thickness of the muscle measuring system according to claim 6, characterized in that said system further comprising: a preprocessing module for preprocessing the image of interest, the pretreatment comprises gray-scale transformation of the image of interest and adjust image contrast.
Descripción  traducido del chino

基于超声图像的肌肉厚度测量方法和系统 Muscle thickness measurement method and system based on ultrasound images

技术领域 Technical Field

[0001] 本发明涉及图像处理领域,特别是涉及一种基于超声图像的肌肉厚度测量方法和系统。 [0001] The present invention relates to the field of image processing, and more particularly to a method and muscle thickness measurement system based on ultrasound images.

背景技术 Background

[0002] 骨骼肌的力学特性是和它的结构形态相关的,任何的身体活动和体育运动,都是由骨骼肌的收缩完成的,这直接影响了人体的力量和耐力。 [0002] The mechanical properties of skeletal muscle and its morphology is associated with any physical activity and sports are done by the contraction of skeletal muscles, which directly affects the body's strength and endurance. 肌肉具有一定的弹性,被拉长后,当拉力解除时可自动恢复到原来的程度,肌肉的弹性可以减缓外力对人体的冲击,因而在运动中扮演着至关重要的作用。 Muscle has a certain flexibility, after being stretched when releasing the tension can be automatically restored to its original level, muscle flexibility can slow external shocks on the human body, and thus play a vital role in the movement. 而肌肉的构成又十分复杂,定量分析和评估肌肉功能状态是运动医学和运动功能康复研究中的难点和热点。 The composition of muscle and is very complex, quantitative analysis and evaluation of muscle function is a function of the state of sports medicine and sports rehabilitation research in the difficult and hot.

[0003]目前对于肌肉的厚度的测量,如厚度的测量大部分采取的是人工手动测量,因手动测量对环境等诸多主观因素相当敏感,使得测量缺乏客观性,测量精度难以控制,并且对于测量大批量的肌肉厚度图片,操心过程费时费力,测量效率低。 [0003] At present, for measuring the thickness of the muscle, such as measuring the thickness of the most taken by manually measurement, due to manual measurement on the environment and many other subjective factors are quite sensitive, so measuring lack of objectivity, measurement accuracy is difficult to control, and for measurements Large quantities of muscle thickness picture, worry about time-consuming process, low measuring efficiency. 另外骨骼肌在运动过程中肌肉厚度的变化在每帧图像中比较细微,测量容易失真,从而影响测量结果。 Further changes in the muscles during exercise in skeletal muscle thickness is relatively slight in each frame, the measurement is easy to distortion, thus affecting the measurement results.

发明内容 DISCLOSURE

[0004] 基于此,有必要针对现有技术中测量效率低且测量不准确的问题,提供一种能提高测量准确度和测量效率的基于超声图像的肌肉厚度测量方法。 [0004] Based on this, it is necessary for the low efficiency of the art measurement and the measurement is not accurate, provide a measure to improve the measurement accuracy and efficiency measurement methods based on ultrasound images of muscle thickness.

[0005] 此外,还有必要针对现有技术中`测量效率低且测量不准确的问题,提供一种能提高测量准确度和测量效率的基于超声图像的肌肉厚度测量系统。 [0005] In addition, there is a need in the art for measuring efficiency and low `the problem of inaccurate measurements, provide a measure to improve the measurement accuracy and efficiency of ultrasound images based on muscle thickness measurement systems.

[0006] 一种基于超声图像的肌肉厚度测量方法,包括以下步骤: [0006] A muscle thickness measurement method is based on the ultrasound image, comprising the steps of:

[0007] 从捕捉的超声图像中提取感兴趣图像; [0007] extracted from the captured image of interest ultrasound image;

[0008] 获取在所述感兴趣图像中选择的多个初始跟踪窗口的位置; [0008] to obtain the location of the selected image in the interest of multiple initial tracking window;

[0009] 对多个跟踪窗口进行跟踪,并通过跟踪算法确定后续每帧图像相应的多个跟踪窗口的位置; [0009] The plurality of tracking window to track and identify each subsequent frame image corresponding plurality of tracking algorithm by tracking the position of the window;

[0010] 对每帧图像的多个跟踪窗口中与周围图像模态相似的跟踪窗口采用取对角线交点为中心点进行处理,对每帧图像中其余跟踪窗口采用边缘检测法进行处理; [0010] For each frame of the plurality of tracking window surrounding the image modality similar tracking window using diagonal intersection take center for processing, for each frame remaining trace window treatment using edge detection method;

[0011] 计算每帧图像中经过中心点处理后的跟踪窗口的位置与每个经过边缘检测法处理后的跟踪窗口的位置之间的最大垂直距离,将所述最大垂直距离作为经过中心点处理后的跟踪窗口与经过边缘检测法处理后的跟踪窗口之间的肌肉厚度值。 [0011] calculated for each frame image after processing center position after Trace window after the maximum vertical distance of each position of the trailing edge detection process Trace window between the maximum vertical distance as the center after treatment muscle thickness value of the trace window and trace window after edge detection process between.

[0012] 一种基于超声图像的肌肉厚度测量系统,包括: [0012] Based on ultrasound images of muscle thickness measurement system, comprising:

[0013] 提取模块,用于从捕捉的超声图像中提取感兴趣图像; [0013] The extraction module for extracting the image of interest from the captured ultrasound images;

[0014] 获取模块,用于获取在所述感兴趣图像中选择的多个初始的跟踪窗口的位置; [0014] acquisition module for acquiring the position of the selected image in the interest of the plurality of initial tracking window;

[0015] 跟踪模块,用于对多个跟踪窗口进行跟踪,并通过跟踪算法确定后续每帧图像相应的多个跟踪窗口的位置; [0015] tracking module for tracking multiple windows to track and identify each subsequent frame image corresponding plurality of tracking algorithm by tracking the position of the window;

[0016] 处理模块,用于对每帧图像的多个跟踪窗口中与周围图像模态相似的跟踪窗口采用取对角线交点为中心点进行处理,对每帧图像中其余跟踪窗口采用边缘检测法进行处理; [0016] The processing module for each frame of the plurality of tracking window surrounding the image modality similar tracking window using diagonal intersection take center for processing, for each frame remaining trace window using edge detection Method for processing;

[0017] 计算模块,用于计算每帧图像中经过中心点处理后的跟踪窗口的位置与每个经过边缘检测法处理后的跟踪窗口的位置之间的最大垂直距离,将所述最大垂直距离作为经过中心点处理后的跟踪窗口与经过边缘检测法处理后的跟踪窗口之间的肌肉厚度值。 [0017] The calculation module for calculating the position of each frame image after post-processing of the center point of the trace window after the maximum vertical distance between the position of the edge detection processing between each of the trace window, the maximum vertical distance As muscle thickness values after the trace window after processing center and edge detection process after tracking windows.

[0018] 上述基于超声图像的肌肉厚度测量方法和系统,通过对选取的多个跟踪窗口进行跟踪,通过跟踪算法确定在后续每帧图像中的多个跟踪窗口的位置,计算每帧图像中的经过中心点处理的跟踪窗口的位置与边缘检测法处理后的跟踪窗口的位置之间的最大垂直距离,作为肌肉厚度值,该测量方法基于超声图像,并采用图像算法进行修正处理得到的肌肉厚度值较为准确,提高了测量的准确度以及测量效率,且能跟踪后续每帧图像的跟踪窗口,并测量每帧图像中的肌肉厚度值,达到了实时测量的目的。 [0018] Based on the above ultrasound image muscle thickness measurement method and system for selecting a plurality of tracking through the window track, determine the location of each image in the subsequent multiple trace window by tracking algorithm to calculate each frame image After the maximum vertical distance between the position after position and edge detection processing center processing Trace window Trace window between muscle thickness values as the measuring method based on ultrasound images and using image processing algorithms to correct the resulting muscle thickness value is more accurate, and improve the accuracy and measurement efficiency measurements, and can track each subsequent frame image tracking window and measure the thickness of the muscle in each frame value, achieve the purpose of real-time measurements.

附图说明 Brief Description

[0019] 图1为一个实施例中基于超声图像的肌肉厚度测量方法的流程示意图; [0019] FIG. 1 is a flow muscle thickness measurement method based on the example of a schematic diagram of one embodiment of the ultrasound image;

[0020] 图2为预处理后的超声图像; [0020] FIG. 2 is pretreated ultrasound image;

[0021] 图3为图像中跟踪窗口的界定与精确定位的示意图; [0021] FIG. 3 is a schematic view of the definition and precise positioning of the image tracking window;

[0022] 图4为对多个跟踪窗口进行跟踪,并通过跟踪算法确定后续图像帧的跟踪窗口的位置的流程示意图; [0022] FIG. 4 is a multiple track window track, and a schematic diagram of the process to determine the location of the trace window of subsequent image frames by tracking algorithm;

[0023] 图5为一个实施例中基于超声图像的肌肉厚度测量系统的结构示意图; [0023] FIG. 5 is a block diagram representation of an ultrasound image based on the muscle thickness measurement system according to one embodiment;

[0024] 图6为一个实施例中跟踪模块的内部结构示意图; [0024] FIG. 6 is a diagram illustrating the internal structure of an embodiment of the tracking module;

[0025] 图7为另一个实施例中基于超声图像的肌肉厚度测量系统的结构示意图。 [0025] FIG. 7 is a block diagram representation of an ultrasound image based on the muscle thickness measurement system according to another embodiment.

具体实施方式 DETAILED DESCRIPTION

[0026] 下面结合具体的实施例及附图对基于超声图像的肌肉厚度测量方法和系统的技术方案进行详细的描述,以使其更加清楚。 [0026] The following examples and with reference to specific drawings ultrasound images based technology solutions muscle thickness measurement method and system are described in detail in order to make it more clear.

[0027] 如图1所示,在一个实施例中,一种基于超声图像的肌肉厚度测量方法,包括以下步骤: [0027] Figure 1, in one embodiment, a method of measuring muscle thickness based on ultrasound images, comprising the steps of:

[0028] 步骤S110,从超声图像中提取感兴趣图像。 [0028] step S110, the extracted image of interest from the ultrasound image.

[0029] 本实施例中,通过实时B型超声波扫描仪与一个电子线阵探头获取肌肉的超声图像。 Ultrasound images [0029] In this embodiment, the muscles and get a real-time electronic linear array probe B-type ultrasonic scanner. 具体的,超声波探头的长轴方向垂直地被安排在实验者的大腿上,放置于约40%膝盖的长轴距离处。 Specifically, the long axis direction of the ultrasonic probe is arranged perpendicular to the experimenter's thigh, knee placed in about 40% of the major axis distance. 运用大量的超声凝胶确保探头与皮肤在肌肉收缩期间是声耦合的,调整探头以最优化对比度显示超声图像中的肌肉束。 The use of a large number of ultrasound gel to ensure that the probe and the skin during muscle contraction acoustic coupling, adjust the probe to optimize contrast display an ultrasound image of the muscle bundles. 采用B型超声波扫描仪获取超声图像并传送到视频捕获卡,由其进行数字化处理,并以速度约25帧/秒的采样率采集到计算机内数字化图像采集卡。 Type-B ultrasonic scanners to obtain ultrasound images and to the video capture card, its digitized, and at a rate of about 25 frames / sec sampling rate, to your computer within the digital frame grabbers.

[0030] 对捕捉的超声图像进行裁剪得到感兴趣图像。 [0030] The capture ultrasound images get cropped image of interest. 感兴趣图像即为包含有所需测量的肌肉厚度信息的图像。 We are interested in the image with the image information of the desired muscle thickness measured.

[0031 ] 在Iv实施例中,在从捕捉的超声图像中提取感兴趣图像的步骤之后,还包括步骤:对该感兴趣图像进行预处理,包括:对所述感兴趣图像进行灰度变换及调整图像对比度。 [0031] In Iv embodiment, after the step of extracting the ultrasound image of interest from the captured images, further comprising the step of: preprocessing the image of interest, comprising: gray-scale transformation of the image of interest and Adjust the image contrast. 如图2所示为预处理后的超声图像。 Ultrasound image is shown in Figure 2 after pretreatment. [0032] 步骤S120,获取在该感兴趣图像中选择的多个初始的跟踪窗口的位置。 [0032] step S120, acquires the position of interest in the image, select the trace window of the plurality of initial.

[0033] 具体的,首先手动在感兴趣图像中选择多个初始的跟踪窗口的位置。 [0033] Specifically, first manually select the location of the trace window in the plurality of initial interest in the image. 本实施例中,多个为三个,可手动选择三个初始跟踪窗口,分别跟踪股骨、股直肌上部和股直肌下部边界,如图3所示,窗口A、B和C分别表示上述三个初始跟踪窗口。 In this embodiment, a plurality of three, you can manually select three initial tracking window, respectively, tracking the lower part of the femur, the upper rectus femoris and rectus femoris border, as is shown in the window A, B and C represent the three three initial tracking window.

[0034] 步骤S130,对多个跟踪窗口进行跟踪,并通过跟踪算法确定后续图像帧的跟踪窗口的位置。 [0034] step S130, the tracking of multiple windows to track and determine the location of the trace window of subsequent image frames by tracking algorithms.

[0035] 具体的,跟踪算法为压缩跟踪算法、互相关跟踪算法、形心跟踪算法、质心跟踪算法、波门跟踪算法、边缘跟踪算法、区域平衡跟踪算法等。 [0035] Specifically, the tracking algorithm to compress tracking algorithm, correlation tracking algorithm, centroid tracking algorithm, centroid tracking algorithm gate tracking algorithm, edge tracking algorithm, regional balance tracking algorithm. 互相关跟踪算法是基于图像的相似性度量,在当前图像中寻找最接近基准图像模板区域的一种跟踪算法,它对场景图像质量要求不高,不需分割目标和背景,对与选定的跟踪目标图像不相似的其他一切景物不敏感,能跟踪较小的目标以及目标区域的某一特殊部分或对比度比较差的目标,具有较强的局部抗干扰能力。 Correlation tracking algorithm based on similarity measure of the image, looking for a tracking algorithm template region closest to the reference image in the current image, do not ask it scene image quality, without segmentation target and background, and selected all other scene tracking the target image is not similar insensitive to track small targets and target a particular section or area of relatively poor contrast target, with strong local anti-jamming capability. 互相关算法将基准图像在当前图像上以不同的偏移值位置,根据测量两幅图像之间的相关度函数判断跟踪窗口在当前图像中的位置,跟踪窗口是两个图像匹配最好的位置,即相关函数的峰值。 Cross-correlation algorithm reference image on the current image with a different offset value of the position, based on the correlation function measurement between the two images to determine the location of the trace window in the current image, the tracking window is the best location of the two images match The peak that is the correlation function.

[0036] 步骤S140,对每帧图像的多个跟踪窗口中与周围图像模态相似的跟踪窗口采用取对角线交点为中心点进行处理,对每帧图像中其余跟踪窗口采用边缘检测法进行处理。 [0036] step S140, for each frame of the plurality of tracking window surrounding the image modality similar tracking window using diagonal intersection take center for processing, for each frame remaining trace window using edge detection method deal with.

[0037] 其中,与周围图像模态相似是指窗口内的图像与其附近的图像很相似,通常通过先验知识来确定,在本例超声图像中靠近皮肤的那部分模态是相似的。 [0037] where, with the surrounding image modality similar means an image close to its image within the window is very similar, is usually determined by a priori knowledge, in this case close to the skin of the portion of the ultrasound image modes are similar.

[0038] 如图3所示,由于与周围图像模态相似,跟踪窗口A采用取对角线交点为中心点的中心点法进行处理,跟踪窗口B和C采用边缘检测法进行处理,该边缘检测法可为canny算子的边缘检测法。 [0038] 3, due around the image modality similar tracking window A diagonal intersection take center using the center point method processing, tracking window B and C are processed using edge detection method, the edge assay for the canny edge detection method. 采用canny算子的边缘检测将窗口图像变换成为二进制图像,参数被调整以确保获得更多的组织细节,再运用最大连通区域搜索技术寻找每个窗口的确切边界。 Using canny edge detection operator window image converted into binary image, the parameters are adjusted to ensure organizations get more details, then use the largest connected area search technology to find the exact boundaries of each window.

[0039] 步骤S150,计算每帧图像中经过中心点处理后的跟踪窗口的位置与每个经过边缘检测法处理后的跟踪窗口的位置之间的最大垂直距离,将所述最大垂直距离作为经过中心点处理后的跟踪窗口与经过边缘检测法处理后的跟踪窗口之间的肌肉厚度值。 [0039] step S150, to calculate the position of each frame image after processing center trace window after the maximum vertical distance between the position of the edge detection process after tracking window between each, the maximum vertical distance as through muscle thickness values after the trace window and trace window treatment center after edge detection process between.

[0040] 具体的,以如图3中跟踪窗口A、B和C为例,计算每时刻每帧图像中跟踪窗口A与B之间的最大垂直距离,得到股直肌的厚度(Rectus femorisThickness, RFT),跟踪窗口A与C之间的最大垂直距离,得到股四头肌的厚度(QMT )。 [0040] Specifically, in order to track the window shown in Figure 3 A, B and C, for example, to calculate each frame every time track maximum vertical distance between the windows A and B, between obtained rectus femoris thickness (Rectus femorisThickness, The maximum vertical distance RFT), trace window between A and C, get quadriceps thickness (QMT).

[0041] 上述基于超声图像的肌肉厚度测量方法,通过对选取的多个跟踪窗口进行跟踪,通过跟踪算法确定在后续每帧图像中的多个跟踪窗口的位置,计算每帧图像中的经过中心点处理的跟踪窗口的位置与边缘检测法处理后的跟踪窗口的位置之间的最大垂直距离,作为肌肉厚度值,该测量方法基于超声图像,并采用图像算法进行修正处理得到的肌肉厚度值较为准确,提高了测量的准确度以及测量效率,且能跟踪后续每帧图像的跟踪窗口,并测量每帧图像中的肌肉厚度值,达到了实时测量的目的。 [0041] The above-mentioned muscle thickness measurement method based ultrasound images, by selecting a plurality of track window track, determine the location of each image in the subsequent multiple trace window by tracking algorithm to calculate each frame image through the center The maximum vertical distance between the position of the track after the point where the processing window and edge detection process between the trace window, as muscle thickness values, the measurement method is based on ultrasound images and using image correction algorithm processing muscle thickness values obtained more accurate, and improve the accuracy and measurement efficiency measurements, and can track each subsequent frame image tracking window and measure the thickness of the muscle in each frame value, achieve the purpose of real-time measurements.

[0042] 进一步的,在一个实施例中,如图4所示,跟踪算法为压缩跟踪算法。 [0042] Further, in one embodiment, shown in Figure 4, the tracking algorithm compression tracking algorithm. 步骤S130具体为: Step S130 particular:

[0043] 步骤S131,对跟踪窗口所在帧图像进行采样,得到属于跟踪窗口位置范围内的样本集合。 [0043] step S131, where the tracking window frame image is sampled to obtain a sample belonging to the tracking window range of positions within the set.

[0044] 具体的,输入第t帧图像,对t帧图像的一系列图像片段进行采样,依据条件为:[0045] [0044] Specifically, the input image t-th frame, a series of image slices t frame image is sampled, according to the conditions as follows: [0045]

Figure CN103027713AD00071

[0046] 其中,It_i是在第t_l时刻的跟踪位置; [0046] where, It_i is the first time t_l track position;

[0047] r是设定的度量当前图像与It_i之间差别的参数值,r越小,说明当前图像I(Z)与It-1相差越小; [0047] r is a measure of the difference between the image and the current parameter value setting It_i, r is smaller, the current image I (Z) and It-1 difference is smaller;

[0048] Dr是指在所有跟踪的位置中属于跟踪窗口位置范围内的像素点即正样本的集合; [0048] Dr tracking refers to all positions belonging to the set of pixels that is positive samples range of positions within the tracking window;

[0049] I (Z)表示在t时刻获得的跟踪窗口的位置。 [0049] I (Z) indicates the position at time t obtained Trace window.

[0050] 此外,通过采集靠近选择的跟踪窗口的正样本和远离跟踪窗口的负样本对分类器进行更新。 [0050] In addition, through the acquisition of selected trace window near the positive samples and negative samples away from the Trace window classifier updated.

[0051] 对感兴趣的图像可采用较高分辨率进行滤波,对其它部分采取较低分辨率,以提高处理速度。 [0051] interested in a higher resolution image can be used to filter, to take on the rest of the lower resolution to increase the processing speed.

[0052] 步骤S132,采用稀疏矩阵对样本集合中的每个样本进行降维处理,得到压缩特征向量。 [0052] step S132, sparse matrix sample collection to reduce the dimension of each sample to obtain a feature vector compression.

[0053] 具体的,稀疏矩阵为引入的矩阵,如随机投影V=RX,其中,R是一个随机矩阵,Re Rnxm,其中m>n,运用该公式可以将m维的向量X降维到n维的向量,从而达到降维的作用,V即为压缩特征向量。 [0053] Specifically, the sparse matrix is introduced into the matrix, such as random projection V = RX, wherein, R is a random matrix, Re Rnxm, where m> n, using the formula m-dimensional vector can be reduced dimension to X n dimensional vector, so as to achieve the role of dimensionality reduction, V is the feature vector compression. 对于每一个样本z G Rm,它的低维表示为v=( u p. . .,Un)T e Rn,且需满足m >> n。 For each sample z G Rm, its low-dimensional representation of v = (u p..., Un) T e Rn, and the need to meet m >> n.

[0054] 随机矩阵的选择依据如下: [0054] selected on the basis of random matrices as follows:

[0055] 首先选择稳定的投影矩阵,为了确保信号的线性投影能够保持信号的原始结构,投影矩阵必须满足约束等距性(Restricted isometry property,RIP)条件,然后通过原始信号与测量矩阵的乘积获得原始信号的线性投影测量。 [0055] First select stable projection matrix, in order to ensure linear projection signal to maintain the original structure of the signal, an isometric projection matrix must satisfy the constraint of (Restricted isometry property, RIP) conditions, and then multiplied by the original signal and the measurement matrix is obtained measuring linear projection of the original signal. 此处选取的是随机高斯矩阵R,当TijKo, I), RG Rnxm, Here is a random selection Gaussian matrix R, when TijKo, I), RG Rnxm,

Figure CN103027713AD00072

[0057] 其中,!Tij表示该随机矩阵R中的元素,p表示概率值,s = 2或s = 3,此时满足Johnson-Lindenstrauss 定理。 [0057] wherein,! Tij represents the elements in the random matrix R, p represents probability value, s = 2 or s = 3, this time to meet the Johnson-Lindenstrauss theorem.

[0058] 步骤S133,对压缩特征向量采用分类器进行分类。 [0058] step S133, the compressed feature vectors are classified using the classifier.

[0059] 具体的,对压缩特征向量采用朴素贝叶斯分类器分类,且分类器中的条件概率满足高斯正态分布。 [0059] Specifically, the compressed feature vectors, naive Bayesian classifier, conditional probability and the classification of a Gaussian normal distribution.

[0060] 向量中所有元素都被假定为相互独立。 [0060] all the elements of the vector are assumed to be independent. 当p(y = I) = p (y = 0)时,对每一个压缩特征矢量使用朴素贝叶斯分类器分类。 When p (y = I) = p (y = 0) when, for each feature vector compression Naive Bayes classifier. 计算公式如下: Calculated as follows:

W、. ,Ft-,,叫v=1、 W ,., Ft- ,, called the v = 1,

[0061 ] [0061]

Figure CN103027713AD00073

[0062] y G {0, 1}是一个二进制随机变量,用来表示样本标签。 [0062] y G {0, 1} is a binary random variable that shows a sample label. 分类器H(V)中的条件概率p (Vi |y = I)和POil |y = 0)被假定为满足参数为的高斯正态分布,且 Classifier H (V) in the conditional probability p (Vi | y = I) and POil | y = 0) is assumed to meet the parameters of the Gaussian normal distribution, and

[0063] [0063]

Figure CN103027713AD00074

[0064] 步骤S134,从样本集合中进行抽样得到两组图像样本。 [0064] step S134, a sample from the sample collection to obtain two sets of image samples.

[0065] 具体的,抽样满足且满足a < 4 P,抽样条件分别为: [0065] Specifically, the sample meets and satisfies a <4 P, sampling conditions were:

[0066] Da = Iz II(Z)-1t I < a },D5'0 = {z| “| I (z)_It | < ,其中IT 和D5'0为两组图像样本; [0066] Da = Iz II (Z) -1t I <a}, D5'0 = {z | "| I (z) _It | <, where IT and D5'0 image samples into two groups;

[0067] a,(,@是设定的度量当前图像I (Z)与It之间差别的参数值。 [0067] a, (, @ is a measure of the current image I (Z) parameter values and the difference between the set It is.

[0068] 步骤S135,对所述两组图像样本提取哈尔特征,并采用所述分类器迭代得到相邻的后续图像帧的跟踪窗口的位置。 [0068] step S135, for the two sets of image sample extraction Haar features, and the location of the classifier using iterative get adjacent Trace window subsequent image frames.

[0069] 具体的,迭代算法公式为: [0069] Specifically, the iterative algorithm formula is:

[0070] [0070]

Figure CN103027713AD00081

[0071] 采用公式(2)中的参数对分类器迭代更新,得到第t+1帧图像的跟踪窗口的位置和分类器参数。 [0071] The equation (2) the parameters of the classifier iterative update, get the first t + location and classification parameters an image tracking window.

[0072] 如图5所示,在一个实施例中,一种基于超声图像的肌肉厚度测量系统,包括提取模块110、获取模块120、跟踪模块130、处理模块140和计算模块150。 [0072] As shown in Figure 5, in one embodiment, an ultrasound image based on the measurement of the thickness of the muscle system, including extraction module 110, acquisition module 120, a tracking module 130, processing module 140 and the calculation module 150. 其中: Among them:

[0073] 提取模块110用于从捕捉的超声图像中提取感兴趣图像。 [0073] extraction module 110 is used to extract the image of interest from the captured ultrasound images. 本实施例中,通过实时B型超声波扫描仪与一个电子线阵探头获取肌肉的超声图像。 In this embodiment, the muscles get ultrasound images with an electronic linear array probe through real-time B-mode ultrasound scanner. 具体的,超声波探头的长轴方向垂直地被安排在实验者的大腿上,放置于约40%膝盖的长轴距离处。 Specifically, the long axis direction of the ultrasonic probe is arranged perpendicular to the experimenter's thigh, knee placed in about 40% of the major axis distance. 运用大量的超声凝胶确保探头与皮肤在肌肉收缩期间是声耦合的,调整探头以最优化对比度显示超声图像中的肌肉神经束。 The use of a large number of ultrasound gel to ensure that the probe and the skin during muscle contraction acoustic coupling, adjust the probe to optimize contrast display an ultrasound image of the muscle nerve bundles. 采用B型超声波扫描仪获取超声图像并传送到视频捕获卡,由其进行数字化处理,并以速度约25帧/ 秒的采样率采集到计算机内数字化图像采集卡。 Type-B ultrasonic scanners to obtain ultrasound images and to the video capture card, its digitized, and at a rate of about 25 frames / sec sampling rate, to your computer within the digital frame grabbers. 对捕捉的超声图像进行裁剪得到感兴趣图像。 To capture ultrasound images get cropped image of interest. 感兴趣图像即为包含有所需测量的肌肉厚度信息的图像。 We are interested in the image with the image information of the desired muscle thickness measured.

[0074] 获取模块120获取在该感兴趣图像中选择的多个初始的跟踪窗口的位置。 [0074] acquisition module 120 acquires the position of interest in the image, select the trace window of the plurality of initial.

[0075] 具体的,首先手动在感兴趣图像中选择多个初始的跟踪窗口的位置。 [0075] Specifically, first manually select the location of the trace window in the plurality of initial interest in the image.

[0076] 本实施例中,多个为三个,可手动选择三个初始跟踪窗口,分别跟踪股骨、股直肌上部和股直肌下部边界,如图3所示,窗口A、B和C分别表示上述三个初始跟踪窗口。 [0076] In this embodiment, a plurality of three, you can manually select three initial tracking window, respectively, tracking the femur, the upper rectus femoris and rectus femoris lower boundary, shown in Figure 3, the window A, B and C They represent the three initial tracking window.

[0077] 跟踪模块130用于对多个跟踪窗口进行跟踪,并通过跟踪算法确定后续图像帧的跟踪窗口的位置。 [0077] tracking module 130 is used to track multiple tracking window, and subsequent image frames to determine the trace window algorithm by tracking the position.

[0078] 具体的,跟踪算法为压缩跟踪算法、互相关跟踪算法、形心跟踪算法、质心跟踪算法、波门跟踪算法、边缘跟踪算法、区域平衡跟踪算法等。 [0078] Specifically, the tracking algorithm to compress tracking algorithm, correlation tracking algorithm, centroid tracking algorithm, centroid tracking algorithm gate tracking algorithm, edge tracking algorithm, regional balance tracking algorithm. 互相关跟踪算法是基于图像的相似性度量,在当前图像中寻找最接近基准图像模板区域的一种跟踪算法,它对场景图像质量要求不高,不需分割目标和背景,对与选定的跟踪目标图像不相似的其他一切景物不敏感,能跟踪较小的目标以及目标区域的某一特殊部分或对比度比较差的目标,具有较强的局部抗干扰能力。 Correlation tracking algorithm based on similarity measure of the image, looking for a tracking algorithm template region closest to the reference image in the current image, do not ask it scene image quality, without segmentation target and background, and selected all other scene tracking the target image is not similar insensitive to track small targets and target a particular section or area of relatively poor contrast target, with strong local anti-jamming capability. 互相关算法将基准图像在当前图像上以不同的偏移值位置,根据测量两幅图像之间的相关度函数判断跟踪窗口在当前图像中的位置,跟踪窗口是两个图像匹配最好的位置,即相关函数的峰值。 Cross-correlation algorithm reference image on the current image with a different offset value of the position, based on the correlation function measurement between the two images to determine the location of the trace window in the current image, the tracking window is the best location of the two images match The peak that is the correlation function.

[0079] 处理模块140用于对每帧图像的多个跟踪窗口中与周围图像模态相似的跟踪窗口采用取对角线交点为中心点进行处理,对每帧图像中其余跟踪窗口采用边缘检测法进行处理。 [0079] The processing module 140 for each frame of the plurality of tracking window surrounding the image modality similar tracking window using diagonal intersection take center for processing, for each frame remaining trace window using edge detection Method for processing.

[0080] 如图3所示,与周围图像模态相似的跟踪窗口A采用取对角线交点为中心点的中心点法进行处理,跟踪窗口B和C采用边缘检测法进行处理,该边缘检测法可为canny算子的边缘检测法。 [0080] 3, with the surrounding image modality similar tracking window A diagonal intersection take center using the center point method processing, tracking window B and C are processed using edge detection method, the edge detection Method for the canny edge detection method. 采用canny算子的边缘检测将窗口图像变换成为二进制图像,参数被调整以确保获得更多的组织细节,再运用最大连通区域搜索技术寻找每个窗口的确切边界。 Using canny edge detection operator window image converted into binary image, the parameters are adjusted to ensure organizations get more details, then use the largest connected area search technology to find the exact boundaries of each window.

[0081] 计算模块150用于计算经过边缘检测法处理后的后续图像帧的跟踪窗口的位置与经过中心点处理后的初始的跟踪窗口的位置之间的最大垂直距离,将所述最大垂直距离作为肌肉厚度值。 [0081] calculation module 150 for calculating a location tracking through the rear window edge detection processing subsequent image frame and through the maximum vertical distance between the position after the initial treatment center Trace window between the maximum vertical distance As muscle thickness value.

[0082] 具体的,以如图3中跟踪窗口A、B和C为例,计算每时刻每帧图像中跟踪窗口A与B之间的最大垂直距离,得到股直肌的厚度(Rectus femorisThickness, RFT),跟踪窗口A与C之间的最大垂直距离,得到股四头肌的厚度(QMT )。 [0082] Specifically, in order to track the window shown in Figure 3 A, B and C, for example, to calculate each frame every time track maximum vertical distance between the windows A and B, between obtained rectus femoris thickness (Rectus femorisThickness, The maximum vertical distance RFT), trace window between A and C, get quadriceps thickness (QMT).

[0083] 上述基于超声图像的肌肉厚度测量系统,通过对选取的多个跟踪窗口进行跟踪,通过跟踪算法确定在后续每帧图像中的多个跟踪窗口的位置,计算每帧图像中的经过中心点处理的跟踪窗口的位置与边缘检测法处理后的跟踪窗口的位置之间的最大垂直距离,作为肌肉厚度值,该测量方法基于超声图像,并采用图像算法进行修正处理得到的肌肉厚度值较为准确,提高了测量的准确度以及测量效率,且能跟踪后续每帧图像的跟踪窗口,并测量每帧图像中的肌肉厚度值,达到了实时测量的目的。 [0083] Based on the above ultrasound image of muscle thickness measurement system by selecting multiple trace window track, determine the location of each image in the subsequent multiple trace window by tracking algorithm to calculate each frame image through the center The maximum vertical distance between the position of the track after the point where the processing window and edge detection process between the trace window, as muscle thickness values, the measurement method is based on ultrasound images and using image correction algorithm processing muscle thickness values obtained more accurate, and improve the accuracy and measurement efficiency measurements, and can track each subsequent frame image tracking window and measure the thickness of the muscle in each frame value, achieve the purpose of real-time measurements.

[0084] 在一个实施例中,跟踪算法为压缩跟踪算法时,如图6所示,跟踪模块130包括采样模块131、降维模块132、分类模块133、抽样模块134和迭代模块135。 [0084] In one embodiment, the tracking algorithm compression tracking algorithm, as shown in Figure 6, tracking module 130 includes a sampling module 131, dimension reduction module 132, classification module 133, the sampling module 134 and module 135 iterations. 其中: Among them:

[0085] 采样模块131用于对跟踪窗口所在的帧图像进行采样,得到属于跟踪窗口位置范围内的样本集合。 [0085] sampling module 131 is used to frame the image tracking window where sampling to obtain samples within a collection belonging to the tracking window range of positions.

[0086] 具体的,输入第t帧图像,对t帧图像的一系列图像片段进行采样,依据条件为: [0086] Specifically, the input image t-th frame, a series of image slices t frame image is sampled, according to conditions:

[0087] [0087]

Figure CN103027713AD00091

[0088] [0088]

Figure CN103027713AD00092

其中,It^1是在第t_l时刻的跟踪位置; Wherein, It ^ 1 is in the first position t_l time tracking;

[0089] r是设定的度量当前图像与Iw之间差别的参数值,r越小,说明当前图像I(Z)与It-1相差越小 [0089] r is the value of the current set of metrics and the difference between the image Iw, r is smaller, the current image I (Z) with a smaller difference between the It-1

[0090] Dr是指在所有跟踪的位置中属于跟踪窗口位置范围内的像素点即正样本的集合; [0090] Dr tracking refers to all positions belonging to the set of pixels that is positive samples range of positions within the tracking window;

[0091] I (Z)表不在t时刻犾得的跟踪窗口的位直。 Straight - [0091] I (Z) table is not the time t l is obtained by tracking window.

[0092] 此外,通过采集靠近选择的跟踪窗口的正样本和远离跟踪窗口的负样本对分类器进行更新。 [0092] In addition, through the acquisition of selected trace window near the positive samples and negative samples away from the Trace window classifier updated.

[0093] 对感兴趣的图像可采用较高分辨率进行滤波,对其它部分采取较低分辨率,以提高处理速度。 [0093] interested in a higher resolution image can be used to filter, to take on the rest of the lower resolution to increase the processing speed.

[0094] 降维模块132用于采用稀疏矩阵对样本集合中的每个样本进行降维处理,得到压缩特征向量。 [0094] dimensionality reduction module 132 for sparse matrix sample collection to reduce the dimension of each sample to give compression feature vectors.

[0095] 具体的,稀疏矩阵为引入的矩阵,如随机投影V=RX,其中,R是一个随机矩阵,Re Rnxm,其中m>n,运用该公式可以将m维的向量X降维到n维的向量,从而达到降维的作用,V即为压缩特征向量。 [0095] Specifically, the sparse matrix is introduced into the matrix, such as random projection V = RX, wherein, R is a random matrix, Re Rnxm, where m> n, using the formula m-dimensional vector can be reduced dimension to X n dimensional vector, so as to achieve the role of dimensionality reduction, V is the feature vector compression. 对于每一个样本z G Rm,它的低维表示为v=( u p. . .,Un)T e Rn,且需满足m >> n。 For each sample z G Rm, its low-dimensional representation of v = (u p..., Un) T e Rn, and the need to meet m >> n.

[0096] 随机矩阵的选择依据如下: [0096] selected on the basis of random matrices as follows:

[0097] 首先选择稳定的投影矩阵,为了确保信号的线性投影能够保持信号的原始结构,投影矩阵必须满足约束等距性(Restricted isometry property,RIP)条件,然后通过原始信号与测量矩阵的乘积获得原始信号的线性投影测量。 [0097] First select stable projection matrix, in order to ensure linear projection signal to maintain the original structure of the signal, an isometric projection matrix must satisfy the constraint of (Restricted isometry property, RIP) conditions, and then multiplied by the original signal and the measurement matrix is obtained measuring linear projection of the original signal. 此处选取的是随机高斯矩阵R,当TijKO, I), RG RnXm, Here is a random selection Gaussian matrix R, when TijKO, I), RG RnXm,

Figure CN103027713AD00101

[0099] 其中,r^-表示该随机矩阵R中的元素,p表示概率值,s = 2或s = 3,此时满足Johnson-Lindenstrauss 定理。 [0099] wherein, r ^ - represents the random matrix R elements, p represents probability value, s = 2 or s = 3, this time to meet the Johnson-Lindenstrauss theorem.

[0100] 分类模块133用于对所述压缩特征向量采用分类器进行分类。 [0100] classification module 133 is used for the compression feature vectors, classifiers for classification.

[0101] 具体的,对压缩特征向量采用朴素贝叶斯分类器分类,且分类器中的条件概率满足高斯正态分布。 [0101] Specifically, the compressed feature vectors, naive Bayesian classifier, conditional probability and the classification of a Gaussian normal distribution.

[0102] 向量中所有元素都被假定为相互独立。 [0102] all the elements of the vector are assumed to be independent. When

Figure CN103027713AD00102

时,对每一个压缩特征矢量使用朴素贝叶斯分类器分类。 When, for each feature vector compression Naive Bayes classifier. 计算公式如下: Calculated as follows:

Figure CN103027713AD00103

[0104] y G {0,1}是一个二进制随机变量,用来表示样本标签。 [0104] y G {0,1} is a binary random variable that shows a sample label. 分类器H(V)中的条件概率P(vjy = I)和PGilIy = 0)被假定为满足参数为 Classifier H (V) in the conditional probability P (vjy = I) and PGilIy = 0) is assumed to meet the parameters

Figure CN103027713AD00104

,的高斯正态分布,且 , Gaussian normal distribution, and

[0105] [0105]

Figure CN103027713AD00105

[0106] 抽样模块134用于从所述样本集合中进行抽样得到两组图像梓本。 [0106] sampling module 134 for two sets of image sample obtained from the sample Azusa this collection.

[0107] 具体的,抽样满足且满足a < ( < 0,抽样条件分别为: [0107] Specifically, the sample meets and satisfies a <(<0, the sampling conditions were:

[0108] [0108]

Figure CN103027713AD00106

,其中Da 和D5'0为两组图像样本;a,40是设定的度量当前图像I (Z)与It之间差别的参数值。 Wherein Da and D5'0 image samples into two groups; a, 40 a measure is to set the current image I (Z) and the parameter value of the difference between the It.

[0109] 迭代模块135用于对所述两组图像样本提取哈尔特征,并采用分类器迭代得到相邻的后续图像帧的跟踪窗口的位置。 [0109] Hal iteration module 135 for extracting features of the image of the two samples, and the use of iterations to get the position classification adjacent Trace window subsequent image frames.

[0110] 具体的,迭代算法公式为: [0110] Specifically, the iterative algorithm formula is:

[0111] [0111]

Figure CN103027713AD00107

[0112] 采用公式(2)中的参数对分类器迭代更新,得到第t+1帧图像的跟踪窗口的位置和分类器参数。 [0112] The equation (2) the parameters of the classifier iterative update, get the first t + location and classification parameters an image tracking window.

[0113] 如图7所示,在一个实施例中,上述基于超声图像的肌肉厚度测量系统,除了包括提取模块110、获取模块120、跟踪模块130、处理模块140和计算模块150,还包括预处理模块160。 [0113] Figure 7, in one embodiment, the ultrasound images based on the above-described thickness measuring muscular system, in addition to including extraction module 110, acquisition module 120, a tracking module 130, processing module 140 and the calculation module 150, further comprising a pre- processing module 160. 其中: Among them:

[0114] 预处理模块160用于对所述感兴趣图像进行预处理,所述预处理包括对所述感兴趣图像进行灰度变换及调整图像对比度。 [0114] pre-processing module 160 is used to preprocess the image of interest, including the interest in the pre grayscale image conversion and adjust the image contrast.

[0115] 以上所述实施例仅表达了本发明的几种实施方式,其描述较为具体和详细,但并不能因此而理解为对本发明专利范围的限制。 [0115] The above examples are only the expression of several embodiments of the present invention, the description is more specific and detailed, but it can not therefore be construed as limiting the scope of the invention patent. 应当指出的是,对于本领域的普通技术人员来说,在不脱离本发明构思的前提下,还可以做出若干变形和改进,这些都属于本发明的保护范围。 It should be noted that those of ordinary skill in the art, in the present invention without departing from the idea of the premise, you can also make a number of modifications and improvements, which belong to the scope of the present invention. 因此,本发明专利的保护范围应以所附权利要求为准。 Accordingly, the scope of the present invention patent protection shall be subject to the appended claims.

Citas de patentes
Patente citada Fecha de presentación Fecha de publicación Solicitante Título
CN1681439A *12 Sep 200312 Oct 2005株式会社日立医药Biological tissue motion trace method and image diagnosis device using the trace method
CN101145688A *23 May 200719 Mar 2008沙诺夫公司;沙诺夫欧洲公司Electrostatic discharge protection structures with reduced latch-up risks
CN101464948A *14 Ene 200924 Jun 2009北京航空航天大学Object identification method for affine constant moment based on key point
US5247938 *11 Ene 199028 Sep 1993University Of WashingtonMethod and apparatus for determining the motility of a region in the human body
Otras citas
Referencia
1 *JIZHOU LI ETC.: "Real-Time Detection of Muscle Thickness Changes during Isometric Contraction from Ultrasonography:A Feasibility Study", 《COMPUTERIZED HEALTHCARE(ICCH),2012 INTERNATIONAL CONFERENCE ON》, 18 December 2012 (2012-12-18)
2 *李敏等: "下颌角弧形截骨术后咬肌厚度的变化", 《中华医学美学美容杂志》, vol. 13, no. 2, 30 April 2007 (2007-04-30)
Citada por
Patente citante Fecha de presentación Fecha de publicación Solicitante Título
CN103679754A *20 Dic 201326 Mar 2014河北汉光重工有限责任公司Real-time target searching and tracking technique
CN106683115A *21 Dic 201617 May 2017中国矿业大学Video tracking method based on spiral vision-motion model
Clasificaciones
Clasificación internacionalA61B8/08
Eventos legales
FechaCódigoEventoDescripción
10 Abr 2013C06Publication
8 May 2013C10Entry into substantive examination
25 May 2016C12Rejection of a patent application after its publication