**对比文件名称**:US2013245986A1_Description_20251229_2141
**目标专利名称**:交通工具进入检测
**本次调用的模型名称**:DeepSeek最新版本
| 技术特征描述以及公开性判断结果 | 对比文件原文引用 | 公开性论述 | ||
| **技术特征A:一种用于事件状态检测的方法**<br>**判断结果:直接公开** | [0003] “Systems and methods herein enable a mobile device to detect that a user is traveling in association with a vehicle based at least on motion data.” <br>[0014] “Techniques are described herein for classifying the motion and/or position state of a mobile device, e.g., to determine that a mobile device is located in a vehicle (e.g., car, bus, train, etc.) using a hidden Markov model (HMM) and associated state machine.” | 对比文件全文描述了一种用于检测移动设备是否处于交通工具(如汽车、公交车)中的方法,这属于“事件状态检测”的一种具体应用。因此,对比文件直接公开了“一种用于事件状态检测的方法”这一技术特征。 | ||
| **技术特征B:在计算设备处接收多个传感器信号**<br>**判断结果:直接公开** | [0021] “One or more motion detectors 12 associated with a mobile device provide raw motion data to a motion state classifier module 18.” <br>[0024] “... the motion state classifier module 18 may also optionally utilize data from one or more additional device sensors 14, such as a GPS receiver, Wi-Fi receiver, an audio input device (e.g., a microphone, etc.), or the like.” <br>[0048] “... the HMM takes as input a set of likelihood values p(x(t)\ | ωi ) for the current time t at times t=1, 2, . . . .” | 对比文件明确描述了从与移动设备关联的一个或多个运动检测器(如加速度计)接收原始运动数据,并可选择性地利用来自其他设备传感器(如GPS、Wi-Fi、音频输入设备)的数据。这些数据即为“多个传感器信号”。因此,该特征被直接公开。 | |
| **技术特征C:在所述计算设备处基于所述多个传感器信号使用人工神经网络针对多个时间区间中的每个时间区间确定子事件状态的概率**<br>**判断结果:隐含公开** | [0026] “As shown in FIG. 2A, features extracted from accelerometer data using feature extraction module 202 are processed using a set of statistical models 206 to compute a set of likelihoods using module 204 for various device states.” <br>[0028] “The likelihoods computed at block 254 may be emission probabilities used in a HMM algorithm and/or module. In other embodiments, a support vector machine (SVM) may be used in block 254 to compute likelihoods.” <br>[0030] “Block 308 may consider computed likelihoods 304 and previous probability distribution over motion states at block 312, and compute a new probability distribution over the motion states...” | 对比文件描述了基于传感器信号(如加速度计数据)提取特征,然后使用统计模型(如高斯马尔可夫模型GMM)、支持向量机(SVM)等计算各种设备状态(如行走、跑步、汽车停止、汽车移动等)的“似然值(likelihoods)”或“概率分布(probability distribution)”。这些设备状态对应于目标专利中的“子事件状态”。计算是针对连续的时间区间(t=1,2,...)进行的。虽然对比文件未明确使用“人工神经网络(ANN)”这一术语,但其使用的统计模型(GMM)、SVM以及整个分类系统(如图2A、2B、3所示的分类器)在功能上等同于用于模式识别和概率估计的机器学习模型,属于人工神经网络的范畴或可被其替代。因此,对于本领域技术人员而言,对比文件隐含公开了“使用人工神经网络针对多个时间区间确定子事件状态的概率”这一技术特征。 | ||
| **技术特征D:以及在所述计算设备处基于所述子事件状态的概率经由状态序列模型来检测事件状态**<br>**判断结果:直接公开** | [0014] “... to determine that a mobile device is located in a vehicle (e.g., car, bus, train, etc.) using a hidden Markov model (HMM) and associated state machine.” <br>[0028] “... the likelihoods computed at block 254 may be emission probabilities used in a HMM algorithm and/or module.” <br>[0033] “Thus, a state machine that captures the above observations may be utilized, and a HMM may be built around it to improve the accuracy of detecting vehicular movement.” <br>[0048] “Regardless of the particular state model used, at each time instant t=1, 2, . . . , the HMM takes as input a set of likelihood values p(x(t)\ | ωi ) for the current time t ... The HMM of some embodiments outputs K posterior values p(ωi \ | x(t-L)) corresponding to the probability of being in each state ωi at time t-L ...” | 对比文件明确且反复地教导了使用隐马尔可夫模型(HMM)这一“状态序列模型”,以前述计算出的各状态(子事件)概率(即似然值,作为HMM的发射概率)作为输入,来最终检测更高层级的“事件状态”——即移动设备是否处于交通工具中(“vehicular movement”或“auto”状态)。HMM的输出即为处于各状态的后验概率,用于做出检测决策。因此,该特征被直接公开。 |
| **技术特征E:对所述事件状态的检测包括:基于对所述概率的确定来检测所述子事件状态的序列。**<br>**判断结果:直接公开** | [0033] “The HMM resides above a lower level classifier, which at each time instant, t=1, 2, . . . , outputs a likelihood value p(x(t)\ | ωi ) for lower level motion states wi . ... When the HMM detects the state is either autoStop or autoMove, it concludes that the user is in the higher level state of auto (vehicular movement).” <br>[0048] “... the HMM takes as input a set of likelihood values p(x(t)\ | ωi ) for the current time t at times t=1, 2, . . . .” <br>[0077] “... a Viterbi algorithm 806 can be utilized to provide a most likely state sequence 808 given the input data ...” | 对比文件描述的HMM检测机制正是基于随时间序列输入的各底层运动状态(即子事件状态,如walk, autoStop, autoMove)的概率(似然值),来推断最可能的状态序列(如使用维特比算法),并基于此序列得出用户是否处于“汽车”这一高层事件状态的结论。例如,进入“汽车”状态需要经历特定的状态序列(如walk -> autoStop -> autoMove)。因此,该特征被直接公开。 |
| **技术特征F:检测所述事件状态包括检测用户进入交通工具。**<br>**判断结果:直接公开** | [0003] “Systems and methods herein enable a mobile device to detect that a user is traveling in association with a vehicle ...” <br>[0014] “... to determine that a mobile device is located in a vehicle (e.g., car, bus, train, etc.) ...” <br>[0033] “... it concludes that the user is in the higher level state of auto (vehicular movement).” <br>[0056] “As a result of robust mobile device driving detection as described above...” | 对比文件所解决的技术问题及其核心内容就是检测用户是否在交通工具中(riding in/driving a vehicle)。这是其方法所要检测的“事件状态”。因此,该特征被直接公开。 | ||
| **技术特征G:所述多个传感器信号包括音频信号和至少一个运动传感器信号。**<br>**判断结果:直接公开** | [0021] “One or more motion detectors 12 associated with a mobile device provide raw motion data to a motion state classifier module 18.” (运动传感器信号)<br>[0024] “... the motion state classifier module 18 may also optionally utilize data from one or more additional device sensors 14, such as ... an audio input device (e.g., a microphone, etc.), or the like.” (音频信号) | 对比文件明确教导了分类器模块可以利用来自运动检测器(如加速度计)的运动数据,以及可选的来自音频输入设备(如麦克风)的数据。这直接对应于“音频信号和至少一个运动传感器信号”。因此,该特征被直接公开。 | ||
| **技术特征H:所述至少一个运动传感器信号包括加速度计传感器信号、陀螺仪传感器信号、或磁力计传感器信号中的至少一者。**<br>**判断结果:隐含公开** | [0021] “In some embodiments, the one or more motion detectors 12 are one or more accelerometers providing accelerometer data.” <br>[0068] “The orientation sensors 928 can include, e.g., one or more accelerometers, gyroscopes (gyros), magnetometers, or the like.” | 对比文件明确将加速度计作为主要的运动检测器。同时,在描述设备组件时,提到了方向传感器可包括加速度计、陀螺仪、磁力计等。本领域技术人员知晓,陀螺仪和磁力计是常见的用于检测运动或方向的传感器,可与加速度计结合或单独使用以提供运动信息。因此,在对比文件公开了使用运动传感器信号(特别是加速度计)的基础上,其所涵盖的“运动传感器信号”隐含包括了加速度计、陀螺仪、磁力计中的至少一种。 | ||
| **技术特征I:所述多个传感器信号包括运动传感器信号。**<br>**判断结果:直接公开** | [0021] “One or more motion detectors 12 associated with a mobile device provide raw motion data to a motion state classifier module 18.” <br>[0022] “Some embodiments may be implemented in a camera equipped with one or more gyros, the camera adapted to detect motion via accelerometer data or other kinds of motion data.” | 对比文件的核心实施例就是基于运动检测器(如加速度计)提供的运动数据来进行状态分类。因此,“多个传感器信号包括运动传感器信号”这一特征被直接公开。 | ||
| **技术特征J:所述多个传感器信号包括音频信号。**<br>**判断结果:直接公开** | [0024] “... the motion state classifier module 18 may also optionally utilize data from one or more additional device sensors 14, such as ... an audio input device (e.g., a microphone, etc.), or the like.” | 对比文件明确指出了分类器模块可以可选地利用来自音频输入设备(如麦克风)的数据,即音频信号。因此,该特征被直接公开。 | ||
| **技术特征K:确定所述子事件状态的概率包括:经由卷积神经网络CNN来确定与所述多个传感器信号中的每个传感器信号相对应的状态概率;以及经由深度神经网络DNN来融合所述状态概率。**<br>**判断结果:未公开** | 对比文件中未出现“卷积神经网络(CNN)”或“深度神经网络(DNN)”的表述。其用于计算状态概率/似然值的模块是“statistical models 206”(如图2A)、“compute likelihoods module 254/304”(可能使用SVM或统计模型),以及用于融合或序列处理的“HMM algorithm 258”或概率分布计算模块308。 | 对比文件描述的技术方案使用了一套不同的模型架构来计算和融合状态概率:它使用统计模型(如GMM)或SVM来计算各状态的初始似然值(相当于单个传感器的状态概率?但未明确区分每个传感器),然后使用HMM(一种状态序列模型,而非深度神经网络DNN)来基于时间序列和状态转移概率进行“融合”与推断。CNN和DNN是特定类型的人工神经网络结构,具有特定的层间连接和参数学习方式。对比文件并未描述使用CNN进行特征提取或状态概率计算,也未描述使用DNN(全连接层)来融合来自不同传感器的中间概率。因此,该技术特征既未被直接公开,也未被隐含公开。 | ||
| **技术特征L:所述CNN包括三个神经网络层。**<br>**判断结果:未公开** | 同上,对比文件未公开CNN,更未提及CNN的层数。 | 由于技术特征K(使用CNN和DNN)未被公开,作为其进一步限定的具体层数特征L自然也未被公开。 | ||
| **技术特征M:所述DNN包括单个神经网络层。**<br>**判断结果:未公开** | 同上,对比文件未公开DNN,更未提及DNN的层数。 | 由于技术特征K(使用CNN和DNN)未被公开,作为其进一步限定的具体层数特征M自然也未被公开。 | ||
| **技术特征N:所述状态序列模型包括隐马尔科夫模型或隐半马尔科夫模型。**<br>**判断结果:直接公开** | [0014] “... using a hidden Markov model (HMM) and associated state machine.” <br>[0028] “In other embodiments, a support vector machine (SVM) may be used in block 254 to compute likelihoods. ... such as a Poisson hidden Markov model, hidden Bernoulli model, and other stochastic models ...” <br>[0033] “... a HMM may be built around it to improve the accuracy of detecting vehicular movement.” | 对比文件的核心技术之一就是使用隐马尔可夫模型(HMM)作为状态序列模型。虽然未明确提及“隐半马尔科夫模型”,但HMM是“隐马尔科夫模型或隐半马尔科夫模型”的上位概念或一种具体类型。公开了HMM即直接公开了该特征。 | ||
| **技术特征O:所述事件状态是基于所述子事件状态的序列和所述子事件状态中的每个子事件状态的历时来检测的。**<br>**判断结果:隐含公开** | [0043] “The extended state machine breaks one or more existing states into a set of consecutive sub-states which the system must pass through before exiting that state. This adds stickiness to the system by reshaping the distribution of the state duration (the amount of time spent in a state before transitioning).” <br>[0047] “Additionally, states may have different numbers of sub-states depending on various factors, such as anticipated time within a given state or the like.” <br>[0077] “The filtering may be based on a probabilistic model (e.g., a HMM) configured to restrict transitions ...”(HMM本身建模了状态序列和驻留时间) | 对比文件明确提到了“状态历时(state duration)”的概念,并通过扩展状态机(引入子状态)来显式地建模状态持续时间的分布,以增加状态粘性,防止快速波动。这体现了在检测事件状态(如是否在车中)时,需要考虑各子事件状态(如autoMove)的持续时间(历时)。同时,HMM本身作为一种时间序列模型,其状态转移概率矩阵和扩展状态机的设计,都隐含了对状态序列和历时信息的利用。因此,该特征被隐含公开。 | ||
| **技术特征P:所述子事件状态的序列包括含15个子事件状态的序列。**<br>**判断结果:未公开** | [0033] “For example, ω1 =walk, ω2 =run, ω3 =sit, ω4 =stand, ω5 =fiddle, ω6 =rest, ω7 =autoStop, ω8 =autoMove.” (共8个状态)<br>[0047] “By way of specific, non-limiting example, 5 sub-states may be used for each of walk, run, sit, stand, fiddle, rest, and autoStop, and 12 sub-states for autoMove.” (子状态用于扩展状态机,并非独立的子事件状态) | 对比文件明确列举了其模型中的底层运动状态(即子事件状态)为8个(walk, run, sit, stand, fiddle, rest, autoStop, autoMove)。虽然提到了扩展状态机中使用更多“子状态(sub-states)”,但这些子状态是用于对前述8个主要状态进行历时建模的内部结构,并非目标专利权利要求中定义的、用于构成事件检测序列的“子事件状态”。因此,对比文件未公开“含15个子事件状态的序列”这一具体数量特征。 | ||
| **技术特征Q:经由所述状态序列模型来确定所述事件状态已发生的置信水平,其中当所述置信水平超过阈值时所述事件状态被检测到。**<br>**判断结果:直接公开** | [0027] “The filtered states are then provided to a confidence test block 210 which removes low confidence decisions...” <br>[0029] “At Test Confidence block 260, a similar check to remove states that fail the confidence check...” <br>[0052] “Results provided in the form of state probabilities can be further improved through the use of confidence testing. For example, at each time t, the driving detection decision may be discarded if the posterior probabilities of the two most probable states are comparable to one another (corresponding to a high degree of uncertainty in the decision). Discarding of decisions is based on a confidence threshold...” | 对比文件明确描述了“置信度测试(confidence test)”模块或步骤,其作用是根据一定的置信度阈值来剔除低置信度的状态决策。这等同于“确定事件状态已发生的置信水平,其中当置信水平超过阈值时事件状态被检测到”(反之,未超过阈值或置信度低时,决策被丢弃)。因此,该特征被直接公开。 | ||
| **技术特征R:从所述多个传感器信号中的每个传感器信号中提取特征,其中所述子事件状态的概率是基于所提取的特征来确定的。**<br>**判断结果:直接公开** | [0026] “As shown in FIG. 2A, features extracted from accelerometer data using feature extraction module 202 are processed using a set of statistical models 206 to compute a set of likelihoods using module 204 for various device states.” <br>[0028] “Other variations may include the feature extraction module 252 extracting most or all types of data received from the accelerometer...” | 对比文件明确记载了“特征提取模块(feature extraction module)”,其从传感器信号(如加速度计数据)中提取特征,然后将这些提取的特征输入统计模型来计算各状态的概率(似然值)。这直接公开了该技术特征。 | ||
| **技术特征S:所述计算设备包括可穿戴设备或移动设备,并且其中,所述多个传感器信号是从所述可穿戴设备或移动设备的传感器接收的。**<br>**判断结果:直接公开** | [0002] “Mobile devices are widespread in today's society. For example, people use cellular phones, smart phones, personal digital assistants, laptop computers, pagers, tablet computers, etc. ...” <br>[0021] “One or more motion detectors 12 associated with a mobile device provide raw motion data ...” | 对比文件通篇针对“移动设备(mobile device)”进行描述,并列举了手机、智能手机、平板电脑等具体例子。这些都属于“移动设备”。传感器(运动检测器)是与该移动设备关联的。虽然未明确提及“可穿戴设备”,但智能手机、智能手表等移动设备本身也可以被视为可穿戴设备的一种形式,且本领域技术人员容易理解其传感器集成特性。因此,该特征被直接公开。 | ||
| **技术特征T:对所述多个传感器信号进行时间同步,其中,对所述子事件状态的概率的确定基于经时间同步的传感器信号。**<br>**判断结果:未公开** | 对比文件未提及对来自不同传感器的信号进行“时间同步”的步骤。其描述的重点在于从传感器(主要是加速度计)接收数据,提取特征,然后进行计算和模型处理。 | 对比文件没有教导或暗示需要对多个传感器信号进行时间同步操作。其技术方案可以基于单个主要传感器(加速度计)的数据流工作,即使使用多个传感器,也未提及同步它们的时间戳是必要步骤。因此,该技术特征未被公开。 | ||
| **技术特征U:对所述经时间同步的传感器信号中的每一者进行重采样以使得所述经时间同步的传感器信号具有相同的采样率,其中对所述子事件状态的概率的确定基于重采样的经时间同步的传感器信号。**<br>**判断结果:未公开** | 对比文件未提及对传感器信号进行“重采样”以达到相同采样率的步骤。 | 由于连时间同步的步骤都未公开,作为其后续步骤的“重采样以统一采样率”自然也更未被提及。对比文件的技术方案不依赖于此特征。 | ||
| **技术特征V:基于对所述事件状态的检测来执行动作。**<br>**判断结果:直接公开** | [0056] “As a result of robust mobile device driving detection as described above, various automated applications are enabled.” <br>[0065] “Thus, the determined state of the mobile device may be used to alter or toggle the operation of one or more applications, for example that are being executed by the mobile device, or system controls or configurations of the mobile device.” | 对比文件在描述了驾驶检测方法后,专门用多个段落列举了基于此检测结果可以启用的各种自动化应用和操作,例如切换用户界面到驾驶模式、呼叫转移等。这明确公开了“基于对事件状态的检测来执行动作”这一特征。 | ||
| **技术特征W:执行所述动作包括以下至少一者:激活导航系统;为用户显示信息;或者向交通工具传送要执行一个或多个动作的指示。**<br>**判断结果:直接公开** | [0056] 列举的自动化应用包括:“5) Enabling of in-car radio service and/or other services such as navigation.”(激活导航系统)<br>“1) Switching of user interface to drive mode.”(为用户显示信息-改变UI)<br>“2) Diversion of voice calls to mail box with appropriate message...” (向呼叫方传送指示)<br>“7) Social network updates (e.g., “Driver is currently driving to . . . ”).”(向社交网络传送信息)<br>[0065] “... may cause a busy status to be set at the user's place of work; or may indicate to the user's child ... that the user is en route ...” (向其他设备或人传送指示) | 对比文件明确列举了基于驾驶检测可以执行的具体动作,其中包括:启用导航服务(激活导航系统)、切换用户界面到驾驶模式(可视为一种显示信息的方式)、以及向外部(如呼叫方、社交网络、家人单位)发送消息或状态更新(相当于向交通工具外部实体传送指示)。虽然未明确说“向交通工具传送”,但向外部实体传送指示在功能上类似。因此,该特征被直接公开。 | ||
| **技术特征X:所述多个传感器信号包括至少两种不同类型的传感器信号。**<br>**判断结果:直接公开** | [0021] “One or more motion detectors 12 associated with a mobile device provide raw motion data ...” (类型1:运动传感器,如加速度计)<br>[0024] “... the motion state classifier module 18 may also optionally utilize data from one or more additional device sensors 14, such as a GPS receiver, Wi-Fi receiver, an audio input device (e.g., a microphone, etc.), or the like.” (类型2:GPS、Wi-Fi、音频等) | 对比文件教导了方法可以利用来自运动检测器(如加速度计)的数据,并且可以可选地利用来自其他类型传感器(如GPS、Wi-Fi、音频设备)的数据。这明确涵盖了使用至少两种不同类型传感器信号(如运动+音频,或运动+GPS)的方案。因此,该特征被直接公开。 |
<<<A>>><<<B>>><<<c>>><<<D>>><<<E>>><<<F>>><<<G>>><<<h>>><<<I>>><<<J>>><<<N>>><<<o>>><<<Q>>><<<R>>><<<S>>><<<V>>><<<W>>><<<X>>>