A portrait examination and identification method based on gait cycle decomposition and multi-phase force analysis

Abstract

The application discloses a portrait inspection and identification method based on gait cycle decomposition and multi-phase force analysis, and comprises the following steps: extracting the skeleton key points of a target object in a target object image set in a comparison video and the confidence score of each skeleton key point; extracting the features of the target object in the target object image set according to the skeleton key points of the target object, and obtaining the target object features, wherein the target object features comprise basic features, gait cycle features, gait stage features and gait parameter features; integrating the basic features, the gait cycle features and the gait stage features to obtain comprehensive features; respectively calculating the similarity of the target object features and the comprehensive features in the comparison video to obtain similarity calculation results and overall similarity calculation results; and obtaining the identity identification result of the target object according to the similarity calculation results and the overall similarity calculation results.

Description

Technical Field

[0001] This invention relates to the field of facial recognition technology, and more specifically, to a facial recognition method based on gait period decomposition and multi-temporal force analysis. Background Technology

[0002] 1. Current Status and Limitations of Facial Recognition Technology

[0003] Currently, facial recognition technology mainly relies on static and dynamic features such as "head morphological characteristics", "facial feature morphological characteristics", "facial feature configuration characteristics", and "facial dynamic characteristics" to conduct examination and identification work.

[0004] However, facial recognition technology has significant limitations in practical applications:

[0005] (1) Poor environmental adaptability

[0006] Obscuring issues: When the face is obscured by items such as masks, hats, and scarves, facial recognition and identification will be severely affected.

[0007] Light sensitivity: The accuracy of testing and identification will decrease under complex lighting conditions such as strong light, backlight, and shadow.

[0008] Angle dependence: When the shooting angle deviates from the frontal angle, the effectiveness of the inspection and identification will be greatly reduced.

[0009] (2) Effects of individual changes

[0010] The influence of makeup: Heavy or special makeup can change the original facial features, thus affecting the test results.

[0011] Impact of facial plastic surgery: After facial plastic surgery, the original features are changed, which can also make testing and identification difficult.

[0012] (3) High image quality requirements

[0013] Resolution dependence: Low-resolution images, due to insufficient detail, can affect the accuracy of testing and identification.

[0014] Noise sensitivity: Quality issues such as image blurring and compression distortion can interfere with the inspection and identification process.

[0015] 2. Current Status and Problems of Gait Examination and Identification Techniques

[0016] Gait, as a behavioral biological characteristic, possesses uniqueness, stability, and difficulty in faking, theoretically holding great application potential. However, existing gait examination and identification techniques suffer from the following key problems:

[0017] (1) Lack of standardized technical specifications

[0018] Lack of unified standards: There is currently a lack of technical standards and operating procedures for gait testing and identification.

[0019] Inconsistent methodology: Different institutions use different technical approaches and judgment criteria, resulting in a lack of consistency in the identification results.

[0020] Incomparable results: The lack of standardized result presentation and evaluation systems makes it difficult to compare the evaluation results of different institutions.

[0021] (2) Insufficient feature extraction accuracy

[0022] Coarse-grained features: Existing methods mainly extract macroscopic features such as step size and step frequency, but lack features that can reflect fine individual differences.

[0023] Loss of temporal information: The dynamic changes of each phase within the gait cycle are not fully considered, resulting in incomplete feature information.

[0024] Missing 3D information: It is mainly based on 2D image analysis and lacks 3D spatial information, which cannot fully reflect gait characteristics.

[0025] (3) Insufficient support from biomechanical theory

[0026] Missing force analysis: The changes in lower limb force during each stage of the gait cycle are not considered, making it difficult to gain a deep understanding of the biomechanical characteristics of gait.

[0027] Unclear physiological mechanisms: The lack of in-depth understanding of the physiological mechanisms of human gait limits the further development of gait testing and identification techniques.

[0028] Insufficient analysis of individual differences: No biomechanical-based quantitative model of individual differences has been established, making it impossible to accurately distinguish the gait characteristics of different individuals.

[0029] (4) Low degree of quantification

[0030] Highly subjective: Existing methods mainly rely on expert experience, lack objective quantitative analysis, and are easily affected by human factors.

[0031] Poor reproducibility: Different operators may obtain different identification results, indicating that the existing methods have poor reproducibility.

[0032] Lack of confidence assessment: There is a lack of quantitative assessment of the reliability of the identification results, making it impossible to accurately determine the accuracy of the identification results.

[0033] Therefore, there is an urgent need to develop a human image examination and identification method based on standardized gait period decomposition and three-dimensional skeleton force analysis to address the shortcomings of existing technologies and improve the scientificity, accuracy, and practicality of gait examination and identification. Summary of the Invention

[0034] To address the shortcomings of existing technologies, this invention provides a method for human image verification and identification based on gait period decomposition and multi-temporal force analysis.

[0035] According to one aspect of the present invention, a method for facial recognition and identification based on gait period decomposition and multi-temporal force analysis is provided, comprising:

[0036] Extract the skeleton key points of the target object in the image set of the comparison video and the confidence score of each skeleton key point;

[0037] Based on the skeleton key points of the target object, feature recognition and extraction are performed on the target object in the target object image set to obtain the target object features, which include basic features, gait cycle features, gait phase features and gait parameter features.

[0038] The basic features, gait cycle features, and gait phase features are integrated to obtain comprehensive features;

[0039] The similarity is calculated by comparing the target object features and comprehensive features in the video to obtain the similarity calculation results and the overall similarity calculation results.

[0040] Based on the similarity calculation results and the overall similarity calculation results, the inspection and identification results of the target object are obtained.

[0041] According to another aspect of the present invention, a facial recognition and identification device based on gait period decomposition and multi-temporal force analysis is provided, comprising:

[0042] The first extraction module is used to extract the skeleton key points of the target object in the image set of the comparison video and the confidence score of each skeleton key point.

[0043] The second extraction module is used to perform feature recognition and extraction on the target object in the target object image set based on the skeleton key points of the target object, and to obtain the target object features, including basic features, gait cycle features, gait phase features and gait parameter features.

[0044] The integration module is used to integrate basic features, gait cycle features, and gait phase features to obtain comprehensive features;

[0045] The calculation module is used to compare the target object features and comprehensive features in the video to calculate the similarity, and obtain the similarity calculation result and the overall similarity calculation result.

[0046] The identification module is used to obtain the identity verification result of the target object based on the similarity calculation result and the overall similarity calculation result.

[0047] According to another aspect of the present invention, a computer-readable storage medium is provided, the storage medium storing a computer program for performing the methods described in any of the above aspects of the present invention.

[0048] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: a processor; a memory for storing executable instructions of the processor; the processor being configured to read the executable instructions from the memory and execute the instructions to implement the method described in any of the preceding aspects of the present invention.

[0049] Therefore, this invention establishes a standardized gait cycle decomposition system, dividing left and right strides and further subdividing time phases. Combined with multi-dimensional feature parameter extraction and dynamic force analysis of the skeleton model, it improves recognition accuracy and reduces the false recognition rate. In terms of application, it is suitable for fields such as judiciary, criminal investigation, and security, providing judicial departments with scientific identification methods and assisting in case solving. Scientifically, it is the first to combine biomechanical theory with gait analysis, realizing multiple new technologies and establishing a standardized analysis system. This invention utilizes various technical means to provide reliable quantitative and qualitative testing methods for gait examination, providing solid technical support for future forensic identification and other fields. Attached Figure Description

[0050] Exemplary embodiments of the present invention can be more fully understood by referring to the following figures:

[0051] Figure 1 This is a flowchart illustrating a human image verification and identification method based on gait period decomposition and multi-temporal force analysis provided by an exemplary embodiment of the present invention.

[0052] Figure 2 This is a schematic diagram of the structure of a human image verification and identification method based on gait period decomposition and multi-temporal force analysis provided by an exemplary embodiment of the present invention;

[0053] Figure 3 This is a flowchart of gait feature extraction provided by an exemplary embodiment of the present invention;

[0054] Figure 4 This is an exemplary embodiment of the present invention, which provides a normal human gait cycle and time phase diagram.

[0055] Figure 5 This is a measurement diagram of step length, stride length, step width, and step angle provided in an exemplary embodiment of the present invention;

[0056] Figure 6 This is a flowchart of similarity calculation provided by an exemplary embodiment of the present invention;

[0057] Figure 7This is a schematic diagram of the structure of a human image verification and identification device based on gait period decomposition and multi-temporal force analysis provided by an exemplary embodiment of the present invention;

[0058] Figure 8 This is the structure of an electronic device provided in an exemplary embodiment of the present invention. Detailed Implementation

[0059] Hereinafter, exemplary embodiments according to the present invention will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments of the present invention. It should be understood that the present invention is not limited to the exemplary embodiments described herein.

[0060] It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0061] Those skilled in the art will understand that the terms "first," "second," etc., in the embodiments of the present invention are only used to distinguish different steps, devices, or modules, and do not represent any specific technical meaning, nor do they indicate a necessary logical order between them.

[0062] It should also be understood that in the embodiments of the present invention, "multiple" can refer to two or more, and "at least one" can refer to one, two or more.

[0063] It should also be understood that any component, data or structure mentioned in the embodiments of the present invention can generally be understood as one or more unless explicitly defined or given contrary instructions in the context.

[0064] Furthermore, the term "and / or" in this invention is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this invention generally indicates that the preceding and following related objects have an "or" relationship.

[0065] It should also be understood that the description of the various embodiments in this invention emphasizes the differences between the various embodiments, and the similarities or similarities can be referred to each other. For the sake of brevity, they will not be described in detail.

[0066] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0067] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the invention or its application or use.

[0068] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, they should be considered part of the specification.

[0069] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0070] The embodiments of this invention can be applied to electronic devices such as terminal devices, computer systems, and servers, and can operate together with a wide range of other general-purpose or special-purpose computing system environments or configurations. Well-known examples of terminal devices, computing systems, environments, and / or configurations suitable for use with electronic devices such as terminal devices, computer systems, and servers include, but are not limited to: personal computer systems, server computer systems, thin clients, thick clients, handheld or laptop devices, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputer systems, mainframe computer systems, and distributed cloud computing environments including any of the above systems, etc.

[0071] Electronic devices such as terminal devices, computer systems, and servers can be described in the general context of computer system executable instructions (such as program modules) executed by a computer system. Typically, program modules can include routines, programs, object programs, components, logic, data structures, etc., which perform specific tasks or implement specific abstract data types. Computer systems / servers can be implemented in distributed cloud computing environments, where tasks are executed by remote processing devices linked through communication networks. In distributed cloud computing environments, program modules can reside on local or remote computing system storage media, including storage devices.

[0072] Exemplary methods

[0073] Figure 1 This is a schematic flowchart of a facial recognition and identification method based on gait period decomposition and multi-temporal force analysis, provided by an exemplary embodiment of the present invention. This embodiment can be applied to electronic devices, such as… Figure 1 As shown, the facial recognition and identification method 100 based on gait period decomposition and multi-temporal force analysis includes the following steps:

[0074] Step 101: Extract the skeleton key points of the target object in the target object image set in the comparison video and the confidence score of each skeleton key point;

[0075] Step 102: Based on the skeleton key points of the target object, perform feature recognition and extraction on the target object in the target object image set to obtain the target object features, which include basic features, gait cycle features, gait phase features and gait parameter features.

[0076] Step 103: Integrate the basic features, gait cycle features, and gait phase features to obtain comprehensive features;

[0077] Step 104: Calculate the similarity of the target object features and the comprehensive features in the comparison video to obtain the similarity calculation results and the overall similarity calculation results.

[0078] Step 105: Based on the similarity calculation results and the overall similarity calculation results, obtain the inspection and identification results of the target object.

[0079] Specifically, addressing the technical problems existing in the background art, this application provides a facial recognition method based on gait period decomposition and multi-temporal force analysis to solve the problems existing in the prior art in facial recognition and achieve high-precision and high-reliability facial recognition. (Reference) Figure 2 As shown, it includes the following steps:

[0080] Step 1: Video Preprocessing

[0081] Load the video of the evidence and the video of the sample; perform format detection, quality assessment and preprocessing on the video; normalize the video frame rate to 30fps; perform noise reduction, enhancement and stabilization on the video.

[0082] Step 2: Human Pose Estimation

[0083] Using a deep learning model, 33 key points of the human body were detected, including:

[0084] Head area (0-10): key points such as nose, eyes, and ears

[0085] Trunk area (11-22): Key points such as shoulders, chest, and waist

[0086] Lower limb area (23-32): Key points such as hip, knee, ankle, and toes.

[0087] Assign a confidence score in the range of 0-1 to each keypoint; track the positional changes of keypoints in the time series; identify and filter anomalous or unreliable pose data.

[0088] Step 3: Gait feature extraction, such as Figure 3 As shown,

[0089] Calculate the fundamental characteristics:

[0090] Foot height: ;

[0091] In the formula, : Foot height at time t. : The vertical coordinate of the heel at time t. : The vertical coordinate of the toe tip at time t. Take the larger of the two values.

[0092] Note: The point of contact between the foot and the ground is the lowest point of the foot. The height of the foot is determined by this lowest point. This calculation method ensures that the height reflects the actual position of the foot.

[0093] Foot speed: ;

[0094] In the formula, Foot velocity at time t, in pixels per second. - Horizontal displacement. - Vertical displacement. Total displacement (Euclidean distance). Time step; usually (fps refers to video frame rate).

[0095] Knee joint angle: ;

[0096] in:

[0097]

[0098]

[0099] In the formula, : The interior angle at the knee joint at time t (representing the angle between the femur and tibia). a: Thigh length (from hip to knee). b: Tibia length (from knee to ankle). c: Straight line length from hip to ankle. Euclidean distance between two points.

[0100]

[0101]

[0102]

[0103] Foot posture angle: .

[0104] In the formula, : Foot posture angle at time t (representing the angle of inclination of the line connecting the heel and toes to the horizontal axis (X-axis)). Numerator: The difference in vertical distance between the toes and the ankle joint. Denominator: The difference in horizontal distance between the toes and ankle. Ratio: The slope of the line connecting the feet.

[0105] Perform gait period modeling, such as Figure 4 As shown:

[0106] Gait cycle definition: a complete gait cycle It includes two strides:

[0107]

[0108] Left stride modeling:

[0109] Right stride modeling:

[0110] Identify 6 gait phases:

[0111] ① First stage (early stage):

[0112] Pre-swing judgment formula

[0113] Judgment conditions:

[0114]

[0115] Formula Analysis: This is a Boolean function that outputs either 0 or 1. It uses a logical AND (∧) to connect all conditions, requiring all conditions to be met simultaneously. It is a piecewise function, where the value is 1 when the condition is met and 0 otherwise.

[0116] Mathematical expression:

[0117]

[0118] Brief analysis of formula:

[0119] It is an indicator function.

[0120]

[0121] Multiplying multiple indicator functions is equivalent to a logical AND operation; the product is 1 only when all conditions are true.

[0122] This indicates that the foot has an upward tendency to move vertically. In the early stages of take-off, the foot begins to prepare to leave the ground, generating upward acceleration. It is the threshold for upward velocity, usually set to 0.1.

[0123] Ensure the ankle joint is positioned high enough to avoid misjudgment. It is the height threshold, usually set to 0.8 (the ratio relative to human height).

[0124] The toes should also be positioned high enough to match the ankle height, ensuring that the entire foot is ready to leave the ground.

[0125] The knee joint is in a relatively extended state, with an angle ranging from 140° to 160°, close to full extension (180°).

[0126] The feet move forward in the horizontal direction, ensuring they are not stationary or moving backward.

[0127] Characteristic function:

[0128]

[0129] in:

[0130] Formula analysis: This is a linear combination function, where each feature has a corresponding weight coefficient, which reflects the importance of each feature.

[0131] ②The second stage (starting foot) includes the following judgment formulas:

[0132] Judgment conditions:

[0133]

[0134] Mathematical expression:

[0135]

[0136] In the formula, This indicates that the foot height is below the low threshold, the foot is close to the ground, and during the take-off phase, the foot is preparing to leave the ground; the height should be relatively low. It is a low height threshold, usually set to 0.7.

[0137] This indicates that the foot has a high upward velocity in the vertical direction. During the take-off phase, the foot quickly leaves the ground, generating a large upward acceleration. This is the high-speed threshold, which is usually set to 0.5.

[0138] This indicates that the toes are significantly lower than the heel. The vertical distance difference between the toes and the heel is greater than the threshold δ, which is the position difference threshold, usually set to 10 pixels.

[0139] This indicates that the knee joint is in a bent position, with an angle of less than 120°, which means that the knee joint is significantly bent. This is the typical knee joint posture during the foot take-off phase.

[0140] Characteristic function:

[0141]

[0142] in:

[0143] Brief analysis of formula:

[0144] This is a linear combination function, where each feature has a corresponding weight coefficient, which reflects the importance of each feature.

[0145] Foot height characteristics The reciprocal of the foot's "degree of lift off the ground" indicates the degree of lift off the ground. The larger the value, the closer the foot is to the ground, and the more consistent it is with the characteristics of a foot take-off.

[0146] Vertical velocity characteristics It directly reflects the movement of the foot; a positive value indicates upward movement, and a negative value indicates downward movement.

[0147] Characteristics of the relative positions of the toes This indicates the height difference between the heel and the toes. A positive value indicates that the heel is higher than the toes, and a negative value indicates the opposite. This positional relationship is formed when the toes begin to lift during the take-off phase.

[0148] Knee joint angle characteristics This indicates the degree of knee flexion. 180° is full extension. The larger the value, the more pronounced the knee flexion.

[0149] ③ Third stage (lower leg crossing):

[0150] Calf Cross Judgment Formula

[0151] Judgment conditions:

[0152]

[0153] Mathematical expression:

[0154]

[0155] Brief analysis of formula:

[0156] This is a linear combination function, where each feature has a corresponding weight coefficient, which reflects the importance of each feature.

[0157] Foot posture angle condition: The angle between the foot posture and the horizontal plane is between 45 and 75 degrees, which indicates that the foot is in a moderate tilt state. This is one of the core characteristics of the lower leg crossing state.

[0158] Knee joint angle condition: A knee joint bending angle between 30 and 120 degrees indicates that the knee joint is in a large bending state, which is one of the important support characteristics of the lower leg crossing state.

[0159] Foot height condition: The foot height reaches its peak, indicating that the foot reaches its highest point in the air. This is one of the key characteristics of the lower leg crossing position.

[0160] The ankle joint is located behind the hip joint, indicating that the lower leg is tilted backward relative to the thigh. This is one of the spatial characteristics of the lower leg crossing position.

[0161] The knee joint is in front of the hip joint, indicating that the thigh is tilted forward relative to the hip joint, which is one of the spatial characteristics of the lower leg crossing position.

[0162] Characteristic function:

[0163]

[0164] in .

[0165] Brief analysis of formula:

[0166] This is a linear combination function, where each feature has a corresponding weight coefficient, which reflects the importance of each feature.

[0167] Foot posture and angle characteristics.

[0168] Knee joint angle characteristics.

[0169] Foot height characteristics.

[0170] Anterior displacement of the knee joint.

[0171] : Posterior displacement of the ankle joint.

[0172] ④ Fourth stage (lower leg vertical):

[0173] Calf Verticality Judgment Formula

[0174] Judgment conditions:

[0175]

[0176] Mathematical expression:

[0177]

[0178]

[0179] Brief analysis of formula:

[0180] Knee joint angle condition: When the knee joint is nearly straight, the lower leg and thigh are almost in a straight line. This is one of the core characteristics of the lower leg being in a vertical position.

[0181] Foot height condition: The feet are in the air, at a certain distance from the ground, indicating that the feet are not touching the ground. This is one of the important characteristics of the lower leg being in a vertical position.

[0182] Foot angle condition: The range of angles in which the foot is close to horizontal indicates that the foot is in a posture ready to land, which is one of the important characteristics of the vertical position of the lower leg.

[0183] Ankle joint forward displacement condition: The ankle joint protrudes forward relative to the hip joint, indicating that the lower leg is tilted forward relative to the thigh. This is the spatial characteristic of the lower leg in a vertical state.

[0184] Characteristic function:

[0185]

[0186] in .

[0187] Brief analysis of formula:

[0188] This is a linear combination function, where each feature has a corresponding weight coefficient, which reflects the importance of each feature.

[0189] Knee joint angle characteristics: the closer the knee joint is to full extension, the larger the characteristic value.

[0190] Foot height characteristics: the higher the foot, the larger the characteristic value.

[0191] Foot posture angle characteristics: the closer the foot is to a horizontal posture, the larger the characteristic value.

[0192] Anterior displacement of the ankle joint is a characteristic feature; the greater the anterior displacement of the ankle joint, the larger the characteristic value.

[0193] ⑤ Fifth stage (early stage of implementation):

[0194] Pre-heel strike judgment formula

[0195] Judgment conditions:

[0196]

[0197] Mathematical expression:

[0198]

[0199]

[0200] Brief analysis of formula:

[0201] Vertical velocity condition: When the foot has a downward velocity in the vertical direction, it indicates that the foot is preparing to land. It is the threshold for downward velocity.

[0202] Heel height requirement: Heel height is within the medium range.

[0203] It shouldn't be too high, nor too low.

[0204] : The relative position of the toes is such that the toes are higher than the heels, forming a posture ready to land. δ is the height difference threshold between the toes and the heels.

[0205] Knee joint angle condition: The knee joint should maintain a certain flexion angle to avoid excessive extension of the knee joint, in order to prepare for the absorption of impact when landing.

[0206] Characteristic function:

[0207]

[0208] in: .

[0209] Brief analysis of formula:

[0210] Vertical velocity characteristic reflects the intensity of the foot's downward movement. The greater the downward velocity, the larger the characteristic value, and it is one of the core characteristics in the early stage of foot landing.

[0211] Foot height characteristic reflects how close the foot is to the ground; the lower the height (the closer to the ground), the larger the characteristic value.

[0212] The relative position of the toes is a characteristic. When the toes are significantly higher than the heels (heel on the ground), the value is positive and increases with the height difference. When the height of the toes and heels is similar (neutral), the value is 1 (maximum value). When the toes are significantly higher than the heels (ball on the ground), the value is positive and increases with the height difference.

[0213] in:

[0214]

[0215] The knee joint angle characteristic reflects the degree of knee flexion; the larger the angle, the larger the characteristic value.

[0216] ⑥ Sixth stage (landing):

[0217] Heel_strike judgment formula

[0218] Judgment conditions:

[0219]

[0220] Landing mode determination function:

[0221]

[0222] Mathematical expression:

[0223]

[0224] Brief analysis of formula:

[0225] Foot height condition: The foot height is less than or equal to the ground threshold, ensuring that the foot is on the ground or very close to the ground.

[0226] The vertical velocity condition is that the foot's velocity in the vertical direction approaches zero, meaning that the vertical motion stops the instant the foot touches the ground.

[0227] Knee joint angle condition: The knee joint is in a state of near full extension, with an angle greater than 160° and close to 180° (full extension).

[0228] : Landing mode conditions, ensuring that the landing mode is one of the three valid modes (heel strike: toes are significantly lower than the ankle joint, ball strike: toes are significantly higher than the ankle joint, neutral mode: toes and ankle joint are at similar heights).

[0229] Characteristic function:

[0230]

[0231] in: .

[0232] Landing mode characteristic function:

[0233]

[0234] in: For heel-to-spot mode threshold, The threshold for the foot strike mode.

[0235] Brief analysis of formula: Foot height characteristics reflect the degree to which the feet are off the ground.

[0236] Vertical velocity characteristics reflect the stability of the foot landing.

[0237] Knee joint angle characteristics reflect the degree of knee extension.

[0238] Confidence characteristic, reflecting the reliability of measurement data.

[0239] Landing pattern characteristics reflect the type and intensity of the landing pattern.

[0240] Gait parameters are calculated (step length, stride length, stride width, stride angle, stride frequency, etc.). Figure 5 (as shown)

[0241] In gait parameter calculation, it is necessary to clearly define the concept and usage of heel coordinates. Heel coordinates are a key reference point in gait analysis, used to calculate important gait parameters such as stride length and stride width.

[0242] Heel coordinates:

[0243] Mathematical expression:

[0244] Left heel coordinates:

[0245]

[0246] Right heel coordinates:

[0247]

[0248] in: Let x be the heel's x-coordinate at time t. Let y be the heel coordinate at time t. The confidence level of the heel at time t.

[0249] Step size calculation

[0250] Step length is defined as the distance between two consecutive foot strikes on the same side.

[0251] Mathematical expression:

[0252]

[0253] in:( ， ) represents the heel coordinates of the current frame. () represents the heel coordinates of the foot that made its first landing on the same side.

[0254] Stride Calculation

[0255] Stride length definition: The distance traveled by one foot from the moment it touches the ground until it touches the ground again.

[0256] Mathematical expression:

[0257]

[0258] in:( ， The heel coordinates at the start of the stride, The heel coordinates at the end of the stride.

[0259] Step width calculation

[0260] Stride width is defined as the lateral distance between the two feet.

[0261] Mathematical expression:

[0262]

[0263] in: Let Y be the Y-coordinate of the left heel. Let Y be the Y-coordinate of the right heel.

[0264] Step angle calculation

[0265] Step angle definition: The angle between the foot and the direction of movement.

[0266] Mathematical expression:

[0267]

[0268] in: Toe coordinates, ( () represents the heel coordinates.

[0269] Step frequency calculation

[0270] Step frequency is defined as the number of steps taken per unit of time.

[0271] Mathematical expression:

[0272]

[0273] in: This represents the total number of steps. Total time (seconds).

[0274] Step 4: Feature Processing

[0275] The Dynamic Time Warping (DTW) algorithm is used to align feature sequences of different lengths; the feature data is normalized to a uniform range; the extracted features are combined into a high-dimensional feature vector; and the most discriminative feature subset is selected and optimized.

[0276] Step 5: Similarity Calculation

[0277] like Figure 6 The similarity of stride length, stride length, stride width, stride angle, and stride frequency is calculated separately, as well as the feature similarity and duration similarity of the six stages of left and right stride length (early step, step, lower leg crossing, lower leg vertical, early step, and step). Finally, the weighted average method is used to calculate the overall similarity.

[0278] Step 6: Output Results

[0279] Generate similarity analysis reports; create visual charts to display the results; determine identity matching results based on similarity thresholds.

[0280] Therefore, this application has the following beneficial technical effects:

[0281] 1. Overcoming the limitations of facial recognition technology:

[0282] This method overcomes the problem of significant drop in recognition rate of traditional facial recognition under conditions such as facial occlusion, disguise, and poor image quality, and provides a facial image verification and identification method that is not affected by facial features.

[0283] 2. Establish a standardized gait analysis system:

[0284] Establish a standardized gait cycle decomposition method based on biomechanical theory to achieve standardized and repeatable gait feature extraction, improve the reliability of inspection and identification, and provide a method for quantitative inspection and identification of gait.

[0285] 3. Enable multi-phase force analysis

[0286] By using skeletal modeling and temporal decomposition, we analyze the force changes of the lower limbs at each stage of the gait cycle; reveal the essential differences in gait movements, and provide more refined individual feature recognition.

[0287] 4. Improve the accuracy of identity verification

[0288] By combining quantitative algorithms with qualitative expert analysis, a multidimensional gait feature vector is constructed to achieve high-precision quantitative verification and identification, thereby improving the scientific rigor of the verification and identification process.

[0289] 5. Expand application scenarios

[0290] Develop identity verification technologies applicable to fields such as forensic identification, criminal investigation, and security monitoring, enabling reliable facial recognition and identification even when facial recognition is limited or the target is disguised.

[0291] 6. Standardization Objectives

[0292] Establish a unified standard for gait cycle decomposition, realize a standardized process for gait feature extraction, and ensure the consistency and repeatability of analysis results.

[0293] 7. Precision Target

[0294] Improve the accuracy of identity recognition, reduce the false recognition rate, and enhance sensitivity to individual differences.

[0295] 8. Practical Objectives

[0296] It adapts to different lighting and shooting conditions, supports multiple application scenarios, and provides visualized analysis results.

[0297] 9. Technological Innovation

[0298] For the first time, gait periodicity decomposition and multi-temporal force analysis were combined to establish a gait feature extraction method based on biomechanics and to develop a gait verification and identification algorithm that combines quantitative and qualitative methods.

[0299] 10. Application Innovation

[0300] Expand the application of gait analysis in the field of identity recognition and provide new identification techniques for forensic identification.

[0301] The ultimate goal of this invention is to establish a scientific, accurate, and reliable method for human image examination and identification through standardized gait cycle decomposition and multi-phase force analysis, providing strong technical support for fields such as forensic identification, criminal investigation, and security monitoring.

[0302] Exemplary device

[0303] Figure 7This is a schematic diagram of the structure of a human image verification and identification device based on gait period decomposition and multi-temporal force analysis provided in an exemplary embodiment of the present invention. Figure 7 As shown, the device 700 includes:

[0304] The first extraction module 710 is used to extract the skeleton key points of the target object in the target object image set in the comparison video and the confidence score of each skeleton key point.

[0305] The second extraction module 720 is used to extract features of the target object in the target object image set based on the skeleton key points of the target object, and obtain the target object features, including basic features, gait cycle features, gait phase features and gait parameter features.

[0306] Integration module 730 is used to integrate basic features, gait cycle features and gait phase features to obtain comprehensive features;

[0307] The calculation module 740 is used to calculate the similarity of the target object features and comprehensive features in the comparison video, and obtain the similarity calculation result and the overall similarity calculation result.

[0308] The identification module 750 is used to obtain the identity identification result of the target object based on the similarity calculation result and the overall similarity calculation result.

[0309] Exemplary electronic devices

[0310] Figure 8 This is the structure of an electronic device provided in an exemplary embodiment of the present invention. For example... Figure 8 As shown, the electronic device 80 includes one or more processors 81 and memory 82.

[0311] The processor 81 may be a central processing unit (CPU) or other form of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.

[0312] The memory 82 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 81 may execute the program instructions to implement the methods of the software programs of the various embodiments of the present invention described above, and / or other desired functions. In one example, the electronic device may also include an input device 83 and an output device 84, these components being interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0313] In addition, the input device 83 may also include, for example, a keyboard, a mouse, etc.

[0314] The output device 84 can output various information to the outside. The output device 84 may include, for example, a display, a speaker, a printer, and a communication network and its connected remote output devices, etc.

[0315] Of course, for the sake of simplicity, Figure 8 Only some of the components of this electronic device relevant to the present invention are shown, omitting components such as buses, input / output interfaces, etc. In addition, the electronic device may include any other suitable components depending on the specific application.

[0316] Exemplary computer program products and computer-readable storage media

[0317] In addition to the methods and apparatus described above, embodiments of the present invention may also be computer program products, which include computer program instructions that, when executed by a processor, cause the processor to perform the steps in the methods according to various embodiments of the present invention described in the "Exemplary Methods" section above.

[0318] The computer program product can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of the present invention. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0319] Furthermore, embodiments of the present invention may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the steps of the methods according to various embodiments of the present invention described in the "Exemplary Methods" section above.

[0320] The computer-readable storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, system, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0321] The basic principles of the present invention have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in the present invention are merely examples and not limitations, and should not be considered as essential features of each embodiment of the present invention. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the present invention to the necessity of employing the aforementioned specific details.

[0322] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For system embodiments, since they largely correspond to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0323] The block diagrams of devices, systems, devices, and systems involved in this invention are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, systems, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0324] The methods and systems of the present invention may be implemented in many ways. For example, they may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order of steps for the methods is for illustrative purposes only, and the steps of the methods of the present invention are not limited to the order specifically described above unless otherwise specifically stated. Furthermore, in some embodiments, the present invention may also be implemented as a program recorded on a recording medium, the program comprising machine-readable instructions for implementing the methods according to the present invention. Thus, the present invention also covers recording media storing programs for performing the methods according to the present invention.

[0325] It should also be noted that in the systems, apparatus, and methods of the present invention, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered equivalents of the present invention. The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the invention. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of the invention. Therefore, the invention is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.

[0326] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the invention to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. A method for facial recognition and identification based on gait period decomposition and multi-temporal force analysis, characterized in that, include: Extract the skeleton key points of the target object in the image set of the comparison video and the confidence score of each skeleton key point; Based on the skeleton key points of the target object, feature recognition and extraction are performed on the target object in the target object image set to obtain target object features, wherein the target object features include basic features, gait cycle features, gait phase features and gait parameter features; The basic features, gait cycle features, and gait phase features are integrated to obtain comprehensive features; Similarity calculations are performed on the target object features and the comprehensive features in the comparison video to obtain similarity calculation results and overall similarity calculation results; Based on the similarity calculation results and the overall similarity calculation results, the quantitative verification and identification results of the target object are obtained; The basic features include: Foot height: ; Foot speed: ; Knee joint angle: ; Foot posture angle: ; In the formula, : Foot height at time t; : The vertical coordinate of the heel at time t; : The vertical coordinate of the toe tip at time t; Let be the coordinates of the ankle joint at time t; Let be the coordinates of the toe joint at time t; Thigh length; Calf length; The straight-line length from the hip joint to the ankle joint; The gait cycle features include a left stride cycle and a right stride cycle, wherein the gait phase features of each cycle include: ① The early stage characteristics, the judgment conditions are: Mathematical expression: Characteristic function: in: ; ② The starting foot characteristic, the judgment condition is: Mathematical expression: Characteristic function: in: ; ③ The characteristic of lower leg crossing, the judgment condition is: Mathematical expression: Characteristic function: in ; ④ Vertical features of the lower leg, the judgment criteria are: Mathematical expression: Characteristic function: in ; ⑤ The judgment condition for landing on the early characteristics is: Mathematical expression: Characteristic function: in: ; ⑥ Foot placement characteristics, the judgment conditions are: Landing mode determination function: Mathematical expression: Characteristic function: in: ; Landing mode characteristic function: in: For heel-to-spot mode threshold, Threshold for foot strike mode; : The coordinates of the ankle joint at time t; : The coordinates of the toes at time t; : The coordinates of the knee joint at time t; : The coordinates of the hip joint at time t; : The velocity vector of the foot at time t; : The acceleration vector of the foot at time t; Height threshold; Speed ​​threshold; Angle threshold; Confidence threshold; Similarity calculations are performed on the target object features and the comprehensive features in the comparison video to obtain similarity calculation results and overall similarity calculation results, including: The feature sequences of the target object features and the comprehensive features of the comparison video are aligned using the dynamic time warping algorithm, and then normalized, combined into a high-dimensional feature vector, and a feature subset is selected. Similarity calculations are performed on the selected feature subsets from the comparison videos to obtain the similarity calculation results and the overall similarity calculation results.

2. The method according to claim 1, characterized in that, It also includes preprocessing the images in the target object image set by performing format detection, quality assessment, standardization, denoising, enhancement, and stabilization.

3. The method according to claim 1, characterized in that, Extract the skeleton key points of the target object from the image set, including: Extract the key points of the head of the target object from the target object image set; Extract the torso key points of the target object from the target object image set. Extract the key points of the lower limbs of the target object from the image set of the target object; Based on the head key points, torso key points, and lower limb key points of the target object, the skeletal key points are obtained.

4. The method according to claim 1, characterized in that, The gait parameter features include: stride length, stride width, stride angle, and stride frequency, wherein... Left heel coordinates And the coordinates of the right heel The expression is: In the formula, Let x be the heel's x-coordinate at time t. Let y be the heel coordinate at time t; The mathematical expression for the step size is: In the formula, ( ， ) represents the heel coordinates of the current frame. () represents the heel coordinates of the foot that made the first impact on the same side; The mathematical expression for the stride length is: In the formula, ( ， The heel coordinates at the start of the stride, The heel coordinates at the end of the stride; The mathematical expression for the step width is: In the formula, Let Y be the Y-coordinate of the left heel. The Y-coordinate of the right heel; The mathematical expression for the step angle is: In the formula, Toe coordinates, ( () represents the heel coordinates; The mathematical expression for the step frequency is: in: This represents the total number of steps. Total time (seconds).

5. A facial recognition and identification device based on gait period decomposition and multi-temporal force analysis, used to implement the method described in any one of claims 1-4, characterized in that, include: The first extraction module is used to extract the skeleton key points of the target object in the image set of the comparison video and the confidence score of each skeleton key point. The second extraction module is used to perform feature recognition and extraction on the target object in the target object image set based on the skeleton key points of the target object, and obtain the target object features, wherein the target object features include basic features, gait cycle features, gait phase features and gait parameter features; The integration module is used to integrate the basic features, gait cycle features, and gait phase features to obtain comprehensive features. The calculation module is used to perform similarity calculations on the target object features and the comprehensive features in the comparison video, respectively, to obtain similarity calculation results and overall similarity calculation results; The identification module is used to obtain the inspection and identification results of the target object based on the similarity calculation results and the overall similarity calculation results.

6. A computer-readable storage medium, characterized in that, The storage medium stores a computer program for performing the method described in any one of claims 1-4.

7. An electronic device, characterized in that, The electronic device includes: processor; Memory used to store the processor's executable instructions; The processor is configured to read the executable instructions from the memory and execute the instructions to implement the method described in any one of claims 1-4.

Images