A distance measurement method and system for laser medical equipment
Through adaptive adjustment of detection threshold and wavelet denoising method, combined with data fusion technology, the problem of small distance changes and errors on specific material surfaces of traditional laser medical equipment is solved, which improves the accuracy and stability of distance measurement, and adapts to Disturbance in the medical environment.
Patent Information
- Application Number
- CN202411234667.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-04
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-09-04
AI Technical Summary
The distance measurement method of traditional laser medical equipment has errors when measuring small distance changes or surfaces of specific materials, and electromagnetic interference and light changes in medical environments affect the stability and accuracy of distance measurement.
Adaptive adjustment detection threshold and improved wavelet denoising method are adopted, combined with the data fusion of multiple ranging results and external interference error factors, and the ranging accuracy is improved through the ranging scheme matching model, laser adaptive threshold model and error compensation model. and stability.
Effectively reduce the impact of noise, improve the accuracy of ranging, adapt to different environmental conditions, ensure the accuracy and stability of ranging, and flexibly and efficiently handle different object characteristics and interference factors.
Smart Images

Figure CN118731958B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of laser distance measurement methods, and in particular to a distance measurement method and system for laser medical equipment. Background Art
[0002] In the field of laser medical equipment, accurate distance measurement capabilities are essential to ensure treatment effectiveness and patient safety. However, traditional distance measurement methods may not be accurate enough in some cases, especially when measuring very small distance changes or on surfaces of specific materials, such as highly reflective or absorbent materials, which may lead to measurement errors. In addition, there may be various electromagnetic interferences, light changes or noise generated by other equipment in the medical environment, which may affect the stability and accuracy of the laser distance measurement system. In order to improve the accuracy of distance measurement, these interference factors need to be effectively processed and compensated. Summary of the invention
[0003] The present invention aims to provide a distance measurement method and system for laser medical equipment to improve the accuracy of distance measurement.
[0004] A distance measurement method for laser medical equipment comprises the following steps:
[0005] S1. Obtain information about an object to be measured, wherein the information about the object to be measured includes a starting point to be measured, an end point to be measured, and feature information of the object to be measured; the feature information of the object to be measured includes image information of the object to be measured and text information of the object to be measured; based on the feature information of the object to be measured and the ranging scheme matching model, an analysis is performed to obtain a ranging strategy for the object to be measured; the ranging strategy for the object to be measured includes an initial detection threshold for ranging of the object to be measured and a ranging accuracy standard for the object to be measured; the ranging accuracy standard for the object to be measured includes the number of times the ranging of the object to be measured is performed; the ranging scheme matching model is used to match ranging strategies of different accuracy according to different features of the object to be measured; the starting point to be measured, the end point to be measured, the initial detection threshold for ranging of the object to be measured, and the ranging accuracy standard for the object to be measured are combined to obtain an initial ranging scheme for the object to be measured;
[0006] S2. Based on the distance measurement accuracy standard of the object to be measured, the object to be measured is measured N times, where N is the specific value of the number of times the object to be measured is measured, and N is determined by the distance measurement accuracy standard of the object to be measured; the initial detection threshold of the object to be measured is recorded as the distance measurement detection threshold Y1 of the object to be measured;
[0007] For the nth distance measurement, n=1, 2, ..., N, the detection threshold Y is adjusted using the distance measurement of the object to be measured. n The distance to be fitted is obtained by measuring the distance of the object to be measured using the distance measurement strategy of the object to be measured. n ; Get ranging laser signal information X n , based on the ranging laser signal information X nAnalyze the ranging laser adaptive threshold model to obtain the distance adjustment detection threshold Y of the object to be measured n+1 ; In the ranging laser adaptive threshold model, the improved wavelet denoising method is used to denoise the laser signal in each ranging, and the denoised signal information and the ranging laser adaptive threshold adjustment function are used to analyze and change the detection threshold to improve the ranging accuracy;
[0008] Traverse the distance measurement process N times until the distance to be fitted L1, the distance to be fitted L2, ..., the distance to be fitted L N ;
[0009] S3. Obtaining external interference information G during N ranging processes according to a preset time interval m , m is the number of times the ranging external interference information is obtained at a preset time interval during the N ranging processes, m=1, 2, ..., M; according to the ranging external interference information G m The error analysis is performed with the ranging error compensation model to obtain the external interference error factor C m ; The ranging error compensation model is used to analyze the external interference information in the ranging process to reduce the error; for all external interference error factors C m Perform data fusion to obtain the comprehensive external error interference factor C'; based on the comprehensive external error interference factor C' and all the distances to be fitted L n Perform data fitting to obtain the distance J of the object to be measured.
[0010] As a preferred technical solution of the present invention, the ranging solution matching model in step S1 includes a data preprocessing layer, a solution matching layer and a result output layer;
[0011] The data preprocessing layer is used to preprocess the feature information of the object to be tested to obtain the image feature information and the text feature information of the object to be tested; the image feature information and the text feature information of the object to be tested are subjected to feature fusion to obtain a fusion feature set of the object to be tested;
[0012] The scheme matching layer is used to perform feature analysis and matching schemes based on the fusion feature set of the object to be measured, and obtain the ranging strategy of the object to be measured;
[0013] The result output layer is used to output the distance measurement strategy of the object to be measured.
[0014] As a preferred technical solution of the present invention, the specific steps of training the scheme matching layer include:
[0015] Collect several groups of historical ranging training samples, each group of historical ranging training samples contains a group of historical ranging object fusion features and corresponding ranging strategies; use the corresponding ranging strategies as target values for a group of historical ranging training samples; combine several groups of historical ranging training samples to obtain ranging scheme matching training sets;
[0016] The ranging scheme matching training set is input into the ranging scheme matching model to train the scheme matching layer with the target value as the target, and an initial scheme matching layer is obtained; a model evaluation is performed on the initial scheme matching layer to obtain an initial scheme matching layer model evaluation result; if the initial scheme matching layer model evaluation result is passed, the initial scheme matching layer is used as the scheme matching layer in the ranging scheme matching model; otherwise, the ranging scheme matching training set is used to continue model training.
[0017] As a preferred technical solution of the present invention, the ranging laser adaptive threshold model in step S2 includes a signal processing layer, a threshold adaptive calculation layer and a threshold output layer;
[0018] The signal processing layer is used to use the improved swarm optimization algorithm to process the ranging laser signal information X n Analyze and get the wavelet denoising parameter setting code B n ; Set the code B according to the wavelet denoising parameters n For ranging laser signal information X n Perform wavelet denoising to obtain the denoised ranging laser signal information Q n ;
[0019] The threshold adaptive calculation layer is used to calculate the denoised ranging laser signal information Q n And ranging laser signal information X n The error value of the ranging laser signal is obtained by n ; Based on the ranging laser signal error value W n , distance measurement of the object to be measured and adjustment of the detection threshold Y n Calculate the distance measurement laser adaptive threshold adjustment function to obtain the distance measurement adjustment detection threshold Y of the object to be measured n+1 ; The ranging laser adaptive threshold adjustment function is set by the historical threshold adjustment information;
[0020] The threshold output layer is used to output the distance measurement of the object to be detected and adjust the detection threshold Y n+1 .
[0021] As a preferred technical solution of the present invention, the improved swarm optimization algorithm is used to calculate the ranging laser signal information X n Analyze and get the wavelet denoising parameter setting code B n The specific steps include:
[0022] Construct K wavelet denoising parameters to set the encoding individual H k , each wavelet denoising parameter setting encodes the individual H k Contains a set of information X for ranging laser signal n The wavelet denoising parameter setting encoding is K; the K wavelet denoising parameter setting encoding individuals H k Combine and get the wavelet denoising parameter setting encoding iteration population; set the maximum number of iterations T max , set the number of iterations t, t=1, 2, …, T max ; The iteration factor is E;
[0023] According to the wavelet denoising parameters, the encoding individual H is set k And ranging laser signal information X n Perform simulation calculations to obtain the wavelet denoising parameter settings for the encoding individual H k The fitness of A k ;
[0024] The specific steps of simulation calculation are:
[0025] Using the formula The calculated wavelet denoising parameters are set to encode individual H k The fitness of A k ; Set the ranging laser signal information X n The length of time is D, X n (d) represents the ranging laser signal information at the dth moment, d = 1, 2, ..., D; U n (d) shows the use of wavelet denoising parameters to set the encoding individual H k The wavelet denoising parameter setting in encodes the simulated denoised ranging laser signal information after denoising at the dth moment;
[0026] Using the formula Calculate the iteration factors E and F max is the maximum dynamic factor, F min is the minimum dynamic factor; when the iteration factor E>1, the wavelet denoising parameter setting encoding iteration population is updated globally; when the iteration factor E<1, the wavelet denoising parameter setting encoding iteration population is updated in the position with larger fitness;
[0027] When the maximum number of iterations T is reached max When the output fitness is the largest, the wavelet denoising parameter setting corresponding to the encoding individual H k , which is the optimal wavelet denoising parameter setting encoding individual; according to the wavelet denoising parameter setting encoding in the optimal wavelet denoising parameter setting encoding individual, the ranging laser signal information X n Perform wavelet denoising.
[0028] As a preferred technical solution of the present invention, the ranging error compensation model in step S3 includes a feature extraction layer, an error compensation factor matching layer and an error compensation factor output layer;
[0029] The feature extraction layer is used to measure the external interference information G m Perform feature analysis to obtain the external interference information feature G of ranging m ';
[0030] The error compensation factor matching layer is used to measure the external interference information characteristics G m 'Perform characteristic analysis and obtain the external interference error factor C m ;
[0031] The error compensation factor output layer is used to output the external interference error factor C m ;
[0032] The specific steps of training the error compensation factor matching layer include:
[0033] Collecting several groups of error factor matching training samples, each group of error factor matching training samples contains a group of external interference information features and target values, and the target value is a corresponding interference error factor; combining several groups of error factor matching training samples to obtain an error compensation factor matching training set;
[0034] The error compensation factor matching training set is input into the ranging error compensation model to train the error compensation factor matching layer with the target value as the target, and an initial error compensation factor matching layer is obtained; a model evaluation is performed on the initial error compensation factor matching layer to obtain an initial error compensation factor matching layer model evaluation result; if the initial error compensation factor matching layer model evaluation result is passed, the initial error compensation factor matching layer is used as the error compensation factor matching layer in the ranging error compensation model; otherwise, the model training is continued using the error compensation factor matching training set.
[0035] As a preferred technical solution of the present invention, based on the comprehensive external error interference factor C' and all the distances to be fitted L n The specific steps for data fitting include:
[0036] Based on all the distances L to be fitted n Calculate the mean and get the mean of the fitting distance L'; use the formula The distance of the object to be measured is calculated, J represents the distance of the object to be measured, and P represents the basic error parameter.
[0037] A distance measuring system for laser medical equipment, comprising:
[0038] The ranging scheme matching module is used to deploy the ranging scheme matching model, obtain the information of the object to be measured, the information of the object to be measured includes the starting point to be measured, the end point to be measured and the feature information of the object to be measured; the feature information of the object to be measured includes the image information of the object to be measured and the text information of the object to be measured; the ranging strategy of the object to be measured is obtained by analyzing the feature information of the object to be measured and the ranging scheme matching model; the ranging strategy of the object to be measured includes the initial detection threshold of the ranging of the object to be measured and the ranging accuracy standard of the object to be measured; the ranging accuracy standard of the object to be measured includes the number of times the ranging of the object to be measured is performed; the ranging scheme matching model is used to match ranging strategies of different accuracy according to different features of the object to be measured; the starting point to be measured, the end point to be measured, the initial detection threshold of the ranging of the object to be measured and the ranging accuracy standard of the object to be measured are combined to obtain the initial ranging scheme of the object to be measured;
[0039] The ranging scheme execution module is used to deploy the ranging laser adaptive threshold model, and perform N distance measurements on the object to be measured based on the ranging accuracy standard of the object to be measured, where N is the specific number of times the distance measurement is performed on the object to be measured; the initial detection threshold of the distance measurement of the object to be measured is recorded as the distance measurement adjustment detection threshold Y1 of the object to be measured; for the nth distance measurement, n=1, 2, ..., N, the detection threshold Y is adjusted using the distance measurement of the object to be measured n The distance to be fitted is obtained by measuring the distance of the object to be measured using the distance measurement strategy of the object to be measured. n ; Get ranging laser signal information X n , based on the ranging laser signal information X n Analyze the ranging laser adaptive threshold model to obtain the distance adjustment detection threshold Y of the object to be measured n+1 ; In the ranging laser adaptive threshold model, the improved wavelet denoising method is used to denoise the laser signal in each ranging, and the denoised signal information and the ranging laser adaptive threshold adjustment function are used to analyze and change the detection threshold to improve the ranging accuracy; the ranging process is traversed N times until the distance to be fitted L1, the distance to be fitted L2, ..., the distance to be fitted L N ;
[0040] The ranging error compensation module is used to deploy the ranging error compensation model and obtain the ranging external interference information G during the N ranging processes according to the preset time interval. m , m is the number of times the ranging external interference information is obtained at a preset time interval during the N ranging processes, m=1, 2, ..., M; according to the ranging external interference information G m The error analysis is performed with the ranging error compensation model to obtain the external interference error factor C m ; The ranging error compensation model is used to analyze the external interference information in the ranging process to reduce the error; for all external interference error factors C mPerform data fusion to obtain the comprehensive external error interference factor C'; based on the comprehensive external error interference factor C' and all the distances to be fitted L n Perform data fitting to obtain the distance of the object to be measured.
[0041] The present invention has the following advantages:
[0042] 1. The present invention can effectively reduce the influence of noise on the ranging result by adaptively adjusting the detection threshold and adopting an improved wavelet denoising method, thereby improving the ranging accuracy; at the same time, external interference information is considered and error compensation is performed, so that the ranging method can maintain high accuracy and stability under different environmental conditions; suitable ranging strategies are matched according to the characteristics of different objects to be measured, making the ranging process more flexible and efficient; by fusing the data of multiple ranging results and external interference error factors, a more reliable comprehensive external error interference factor is obtained, further improving the accuracy of the fitting result.
[0043] 2. The present invention analyzes and processes the ranging laser signal through an improved group optimization algorithm, effectively removes noise and improves the quality of the signal; the group optimization algorithm is used to automatically find the optimal wavelet denoising parameters, which reduces the interference of human factors and improves the denoising effect; through real-time analysis and adjustment, it can adapt to different measurement environments and conditions, improve stability and robustness, and at the same time can dynamically adjust the detection threshold according to the error value of the actual signal, making the ranging method more flexible and accurate. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] Figure 1 The present invention is a schematic diagram of the structure of a distance measurement system of a laser medical device used in an embodiment of the present invention. DETAILED DESCRIPTION
[0045] In order to enable persons skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.
[0046] Embodiment 1, a distance measurement method of a laser medical device, comprising the following steps:
[0047] S1. Obtain information about an object to be measured, wherein the information about the object to be measured includes a starting point to be measured, an end point to be measured, and feature information of the object to be measured; the feature information of the object to be measured includes image information of the object to be measured and text information of the object to be measured; based on the feature information of the object to be measured and the ranging scheme matching model, an analysis is performed to obtain a ranging strategy for the object to be measured; the ranging strategy for the object to be measured includes an initial detection threshold for ranging of the object to be measured and a ranging accuracy standard for the object to be measured; the ranging accuracy standard for the object to be measured includes the number of times the ranging of the object to be measured is performed; the ranging scheme matching model is used to match ranging strategies of different accuracy according to different features of the object to be measured; the starting point to be measured, the end point to be measured, the initial detection threshold for ranging of the object to be measured, and the ranging accuracy standard for the object to be measured are combined to obtain an initial ranging scheme for the object to be measured;
[0048] The information of the object to be measured is collected by sensors, cameras and other equipment. In a medical environment, it is necessary to measure the distance of very small objects. For example, the distance measurement is required before cutting a small steel tube used in surgery, or before positioning with tweezers. If the same distance measurement method is used for all objects, it may lead to increased errors and affect subsequent operations.
[0049] The ranging scheme matching model in step S1 includes a data preprocessing layer, a scheme matching layer and a result output layer;
[0050] The data preprocessing layer is used to preprocess the feature information of the object to be tested to obtain the image feature information and the text feature information of the object to be tested; the image feature information and the text feature information of the object to be tested are subjected to feature fusion to obtain a fusion feature set of the object to be tested;
[0051] The scheme matching layer is used to perform feature analysis and matching schemes based on the fusion feature set of the object to be measured, and obtain the ranging strategy of the object to be measured;
[0052] The result output layer is used to output the distance measurement strategy of the object to be measured;
[0053] The specific steps of training the matching layer include:
[0054] Collect several groups of historical ranging training samples, each group of historical ranging training samples contains a group of historical ranging object fusion features and corresponding ranging strategies; use the corresponding ranging strategies as target values for a group of historical ranging training samples; combine several groups of historical ranging training samples to obtain ranging scheme matching training sets;
[0055] Input the ranging solution matching training set into the ranging solution matching model to train the solution matching layer with the target value as the target, and obtain the initial solution matching layer; perform model evaluation on the initial solution matching layer to obtain the model evaluation result of the initial solution matching layer; if the model evaluation result of the initial solution matching layer is passed, the initial solution matching layer is used as the solution matching layer in the ranging solution matching model; otherwise, continue model training using the ranging solution matching training set;
[0056] Through feature fusion and precise distance measurement strategy matching, the distance measurement accuracy of laser medical equipment under different materials and conditions can be improved; the fusion of image and text feature information can adapt to a more diverse range of objects to be measured and improve the recognition and processing capabilities of different features; through historical data training and model evaluation, the distance measurement strategy can be optimized to more accurately match the actual characteristics of the object to be measured; automated distance measurement strategy matching reduces errors caused by human operation and improves the consistency and reliability of measurement results;
[0057] S2. Based on the distance measurement accuracy standard of the object to be measured, the object to be measured is measured N times, where N is the specific value of the number of times the object to be measured is measured, and N is determined by the distance measurement accuracy standard of the object to be measured; the initial detection threshold of the object to be measured is recorded as the distance measurement detection threshold Y1 of the object to be measured;
[0058] For the nth distance measurement, n=1, 2, ..., N, the detection threshold Y is adjusted using the distance measurement of the object to be measured. n The distance to be fitted is obtained by measuring the distance of the object to be measured using the distance measurement strategy of the object to be measured. n ; Get ranging laser signal information X n , based on the ranging laser signal information X n Analyze the ranging laser adaptive threshold model to obtain the distance adjustment detection threshold Y of the object to be measured n+1 ; In the ranging laser adaptive threshold model, the improved wavelet denoising method is used to denoise the laser signal in each ranging, and the denoised signal information and the ranging laser adaptive threshold adjustment function are used to analyze and change the detection threshold to improve the ranging accuracy;
[0059] Traverse the distance measurement process N times until the distance to be fitted L1, the distance to be fitted L2, ..., the distance to be fitted L N ;
[0060] First, the initial distance measurement scheme of the object to be measured is used to measure the distance of the object to be measured. N is the number of distance measurements set in the distance measurement accuracy standard of the object to be measured. Different accuracy standards are set for different distance measurement objects, which can reduce errors in multiple measurements and improve the accuracy of distance measurement. Through multiple measurements combined with adaptive threshold adjustment, the random error of a single measurement can be effectively reduced and the overall distance measurement accuracy can be improved. The adaptive threshold model can dynamically adjust the detection threshold according to the result of each measurement, thereby compensating for the error in real time. The detection threshold is the lowest threshold intensity that can be received by the laser receiving end. If the noise in the signal is too large, the threshold is adjusted to adapt to the change of noise in the signal.
[0061] The ranging laser adaptive threshold model in step S2 includes a signal processing layer, a threshold adaptive calculation layer and a threshold output layer;
[0062] The signal processing layer is used to use the improved swarm optimization algorithm to process the ranging laser signal information X n Analyze and get the wavelet denoising parameter setting code B n ; Set the code B according to the wavelet denoising parameters n For ranging laser signal information X n Perform wavelet denoising to obtain the denoised ranging laser signal information Q n ;
[0063] The threshold adaptive calculation layer is used to calculate the denoised ranging laser signal information Q n And ranging laser signal information X n The error value of the ranging laser signal is obtained by n ; Based on the ranging laser signal error value W n , distance measurement of the object to be measured and adjustment of the detection threshold Y n Calculate the distance measurement laser adaptive threshold adjustment function to obtain the distance measurement adjustment detection threshold Y of the object to be measured n+1 ; The ranging laser adaptive threshold adjustment function is set by historical threshold adjustment information; The ranging laser adaptive threshold adjustment function collects historical data, uses mathematical methods to fit the threshold adjustment function, and is deployed after verification by professional technicians;
[0064] The threshold output layer is used to output the distance measurement of the object to be detected and adjust the detection threshold Y n+1 ;
[0065] The improved swarm optimization algorithm is used to calculate the ranging laser signal information X n Analyze and get the wavelet denoising parameter setting code B n The specific steps include:
[0066] Construct K wavelet denoising parameters to set the encoding individual H k, each wavelet denoising parameter setting encodes the individual H k Contains a set of information X for ranging laser signal n The wavelet denoising parameter setting encoding is K; the K wavelet denoising parameter setting encoding individuals H k Combine and get the wavelet denoising parameter setting encoding iteration population; set the maximum number of iterations T max , set the number of iterations t, t=1, 2, …, T max ; The iteration factor is E;
[0067] According to the wavelet denoising parameters, the encoding individual H is set k And ranging laser signal information X n Perform simulation calculations to obtain the wavelet denoising parameter settings for the encoding individual H k The fitness of A k ;
[0068] The specific steps of simulation calculation are:
[0069] Using the formula The calculated wavelet denoising parameters are set to encode individual H k The fitness of A k ; Set the ranging laser signal information X n The length of time is D, X n (d) represents the ranging laser signal information at the dth moment, d = 1, 2, ..., D; U n (d) shows the use of wavelet denoising parameters to set the encoding individual H k The wavelet denoising parameter setting in encodes the simulated denoised ranging laser signal information after denoising at the dth moment;
[0070] Using the formula Calculate the iteration factors E and F max is the maximum dynamic factor, F min is the minimum dynamic factor; when the iteration factor E>1, the wavelet denoising parameter setting encoding iteration population is updated globally; when the iteration factor E<1, the wavelet denoising parameter setting encoding iteration population is updated in the position with larger fitness;
[0071] When the maximum number of iterations T is reached max When the output fitness is the largest, the wavelet denoising parameter setting corresponding to the encoding individual H k , which is the optimal wavelet denoising parameter setting encoding individual; according to the wavelet denoising parameter setting encoding in the optimal wavelet denoising parameter setting encoding individual, the ranging laser signal information X n Perform wavelet denoising processing;
[0072] Through the group optimization algorithm, the best wavelet denoising parameters can be found to improve the denoising effect of the ranging laser signal. The wavelet denoising algorithm contains multiple parameter settings. The wavelet denoising algorithm with different parameter settings in the ranging process can adapt to different signal characteristics and noise environments, and provide customized denoising solutions for various ranging scenarios; the improved algorithm can remove noise more effectively and retain useful signal components to adjust the detection threshold;
[0073] S3. Obtaining external interference information G during N ranging processes according to a preset time interval m , m is the number of times the ranging external interference information is obtained at a preset time interval during the N ranging processes, m=1, 2, ..., M; according to the ranging external interference information G m The error analysis is performed with the ranging error compensation model to obtain the external interference error factor C m ; The ranging error compensation model is used to analyze the external interference information in the ranging process to reduce the error; for all external interference error factors C m Perform data fusion to obtain the comprehensive external error interference factor C'; based on the comprehensive external error interference factor C' and all the distances to be fitted L n Perform data fitting to obtain the distance J of the object to be measured;
[0074] External interference information includes temperature changes, fine dust or particles in a non-vacuum environment, etc. The preset time interval is set by professional technicians based on actual conditions;
[0075] The ranging error compensation model in step S3 includes a feature extraction layer, an error compensation factor matching layer and an error compensation factor output layer;
[0076] The feature extraction layer is used to measure the external interference information G m Perform feature analysis to obtain the external interference information feature G of ranging m ';
[0077] The error compensation factor matching layer is used to measure the external interference information characteristics G m 'Perform characteristic analysis and obtain the external interference error factor C m ;
[0078] The error compensation factor output layer is used to output the external interference error factor C m ;
[0079] The specific steps of training the error compensation factor matching layer include:
[0080] Collecting several groups of error factor matching training samples, each group of error factor matching training samples contains a group of external interference information features and target values, and the target value is a corresponding interference error factor; combining several groups of error factor matching training samples to obtain an error compensation factor matching training set;
[0081] The error compensation factor matching training set is input into the ranging error compensation model to train the error compensation factor matching layer with the target value as the target, and obtain the initial error compensation factor matching layer; the initial error compensation factor matching layer is model evaluated to obtain the initial error compensation factor matching layer model evaluation result; if the initial error compensation factor matching layer model evaluation result is passed, the initial error compensation factor matching layer is used as the error compensation factor matching layer in the ranging error compensation model; otherwise, the model training is continued using the error compensation factor matching training set;
[0082] Based on the comprehensive external error interference factor C' and all the distances to be fitted L n The specific steps for data fitting include:
[0083] Based on all the distances L to be fitted n Calculate the mean and get the mean of the fitting distance L'; use the formula The distance of the object to be measured is calculated, J represents the distance of the object to be measured, and P represents the basic error parameter. P is set by professional technicians according to actual conditions;
[0084] By fitting data based on the external error interference factor and the distance to be fitted, the distance of the object to be measured can be calculated more accurately; the application of feature extraction and error compensation factor matching layer helps to identify and compensate for external interference in the ranging process, thereby reducing measurement errors; the ranging error compensation model can dynamically adjust the error compensation strategy according to different external interference information, thereby improving the adaptability of the ranging method; through feature extraction and error compensation, the ranging data can be processed more effectively, improving the quality and efficiency of data processing.
[0085] Example 2, a distance measurement system for laser medical equipment, see Figure 1 As shown, including:
[0086] The ranging scheme matching module is used to deploy the ranging scheme matching model, obtain the information of the object to be measured, the information of the object to be measured includes the starting point to be measured, the end point to be measured and the feature information of the object to be measured; the feature information of the object to be measured includes the image information of the object to be measured and the text information of the object to be measured; the ranging strategy of the object to be measured is obtained by analyzing the feature information of the object to be measured and the ranging scheme matching model; the ranging strategy of the object to be measured includes the initial detection threshold of the ranging of the object to be measured and the ranging accuracy standard of the object to be measured; the ranging accuracy standard of the object to be measured includes the number of times the ranging of the object to be measured is performed; the ranging scheme matching model is used to match ranging strategies of different accuracy according to different features of the object to be measured; the starting point to be measured, the end point to be measured, the initial detection threshold of the ranging of the object to be measured and the ranging accuracy standard of the object to be measured are combined to obtain the initial ranging scheme of the object to be measured;
[0087] The ranging scheme execution module is used to deploy the ranging laser adaptive threshold model, and perform N distance measurements on the object to be measured based on the ranging accuracy standard of the object to be measured, where N is the specific number of times the distance measurement is performed on the object to be measured; the initial detection threshold of the distance measurement of the object to be measured is recorded as the distance measurement adjustment detection threshold Y1 of the object to be measured; for the nth distance measurement, n=1, 2, ..., N, the detection threshold Y is adjusted using the distance measurement of the object to be measured n The distance to be fitted is obtained by measuring the distance of the object to be measured using the distance measurement strategy of the object to be measured. n ; Get ranging laser signal information X n , based on the ranging laser signal information X n Analyze the ranging laser adaptive threshold model to obtain the distance adjustment detection threshold Y of the object to be measured n+1 ; In the ranging laser adaptive threshold model, the improved wavelet denoising method is used to denoise the laser signal in each ranging, and the denoised signal information and the ranging laser adaptive threshold adjustment function are used to analyze and change the detection threshold to improve the ranging accuracy; the ranging process is traversed N times until the distance to be fitted L1, the distance to be fitted L2, ..., the distance to be fitted L N ;
[0088] The ranging error compensation module is used to deploy the ranging error compensation model and obtain the ranging external interference information G during the N ranging processes according to the preset time interval. m , m is the number of times the ranging external interference information is obtained at a preset time interval during the N ranging processes, m=1, 2, ..., M; according to the ranging external interference information G m The error analysis is performed with the ranging error compensation model to obtain the external interference error factor C m ; The ranging error compensation model is used to analyze the external interference information in the ranging process to reduce the error; for all external interference error factors C mPerform data fusion to obtain the comprehensive external error interference factor C'; based on the comprehensive external error interference factor C' and all the distances to be fitted L n Perform data fitting to obtain the distance of the object to be measured.
[0089] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the appended claims of the present invention. Parts not described in detail in this specification belong to the prior art known to those skilled in the art.
Claims
1. A distance measurement method for laser medical equipment, characterized in that: The following steps are involved: S1. Obtain information about the object to be measured, which includes the starting point to be measured, the end point to be measured, and feature information of the object to be measured; The feature information of the object to be measured includes the image information of the object to be measured and the text information of the object to be measured; the distance measurement strategy of the object to be measured is obtained by analyzing the feature information of the object to be measured and the distance measurement scheme matching model; The distance measurement strategy of the object to be measured includes the initial detection threshold of the distance measurement of the object to be measured and the distance measurement accuracy standard of the object to be measured; the distance measurement accuracy standard of the object to be measured includes the number of times the distance measurement of the object to be measured is performed; The ranging scheme matching model is used to match ranging strategies of different precisions according to different features of the object to be measured; the starting point to be measured, the end point to be measured, the initial detection threshold of the distance measurement of the object to be measured and the ranging precision standard of the object to be measured are combined to obtain the initial ranging scheme of the object to be measured; S2. Based on the distance measurement accuracy standard of the object to be measured, the object to be measured is measured N times, where N is the specific value of the number of times the object to be measured is measured, and N is determined by the distance measurement accuracy standard of the object to be measured; The initial detection threshold of the distance measurement of the object to be measured is recorded as the distance measurement adjustment detection threshold Y1 of the object to be measured; For the nth distance measurement, n=1, 2, ..., N, the detection threshold Y is adjusted using the distance measurement of the object to be measured. n The distance to be fitted is obtained by measuring the distance of the object to be measured using the distance measurement strategy of the object to be measured. n ; Get ranging laser signal information X n , based on the ranging laser signal information X n Analyze the ranging laser adaptive threshold model to obtain the distance adjustment detection threshold Y of the object to be measured n+1 ; In the ranging laser adaptive threshold model, the improved wavelet denoising method is used to denoise the laser signal in each ranging, and the denoised signal information and the ranging laser adaptive threshold adjustment function are used to analyze and change the detection threshold to improve the ranging accuracy; Traverse the distance measurement process N times until the distance to be fitted L1, the distance to be fitted L2, ..., the distance to be fitted L N ; S3. Obtaining external interference information G during N ranging processes according to a preset time interval m , m is the number of times the ranging external interference information is obtained at a preset time interval during the N ranging processes, m=1, 2, …, M; According to the external interference information G m The error analysis is performed with the ranging error compensation model to obtain the external interference error factor C m ; The ranging error compensation model is used to analyze the external interference information in the ranging process to reduce the error; for all external interference error factors C m Perform data fusion to obtain the comprehensive external error interference factor C'; based on the comprehensive external error interference factor C' and all the distances to be fitted L n Perform data fitting to obtain the distance J of the object to be measured.
2. A distance measurement method for laser medical equipment according to claim 1, characterized in that: The ranging scheme matching model in step S1 includes a data preprocessing layer, a scheme matching layer and a result output layer; The data preprocessing layer is used to preprocess the feature information of the object to be tested to obtain the image feature information and text feature information of the object to be tested; Performing feature fusion on the image feature information of the object to be tested and the text feature information of the object to be tested to obtain a fused feature set of the object to be tested; The scheme matching layer is used to perform feature analysis and matching schemes based on the fusion feature set of the object to be measured, and obtain the ranging strategy of the object to be measured; The result output layer is used to output the distance measurement strategy of the object to be measured.
3. A distance measurement method for laser medical equipment according to claim 2, characterized in that: The specific steps of training the matching layer include: Collect several groups of historical ranging training samples, each group of historical ranging training samples contains a group of historical ranging object fusion features and corresponding ranging strategies; use the corresponding ranging strategies as target values for a group of historical ranging training samples; combine several groups of historical ranging training samples to obtain ranging scheme matching training sets; The ranging scheme matching training set is input into the ranging scheme matching model to train the scheme matching layer with the target value as the target, and an initial scheme matching layer is obtained; a model evaluation is performed on the initial scheme matching layer to obtain an initial scheme matching layer model evaluation result; if the initial scheme matching layer model evaluation result is passed, the initial scheme matching layer is used as the scheme matching layer in the ranging scheme matching model; otherwise, the ranging scheme matching training set is used to continue model training.
4. The distance measurement method of laser medical equipment according to claim 3, characterized in that: The ranging laser adaptive threshold model in step S2 includes a signal processing layer, a threshold adaptive calculation layer and a threshold output layer; The signal processing layer is used to use the improved swarm optimization algorithm to process the ranging laser signal information X n Analyze and get the wavelet denoising parameter setting code B n ; Set the code B according to the wavelet denoising parameters n For ranging laser signal information X n Perform wavelet denoising to obtain the denoised ranging laser signal information Q n ; The threshold adaptive calculation layer is used to calculate the denoised ranging laser signal information Q n And ranging laser signal information X n The error value of the ranging laser signal is obtained by n ; Based on the ranging laser signal error value W n , distance measurement of the object to be measured and adjustment of the detection threshold Y n Calculate the distance measurement laser adaptive threshold adjustment function to obtain the distance measurement adjustment detection threshold Y of the object to be measured n+1 ; The ranging laser adaptive threshold adjustment function is set by the historical threshold adjustment information; The threshold output layer is used to output the distance measurement of the object to be detected and adjust the detection threshold Y n+1 .
5. The distance measurement method of laser medical equipment according to claim 4, characterized in that: The improved swarm optimization algorithm is used to calculate the ranging laser signal information X n Analyze and get the wavelet denoising parameter setting code B n The specific steps include: Construct K wavelet denoising parameters to set the encoding individual H k , each wavelet denoising parameter setting encodes the individual H k Contains a set of information X for ranging laser signal n The wavelet denoising parameter setting encoding is K; the K wavelet denoising parameter setting encoding individuals H k Combine and get the wavelet denoising parameter setting encoding iteration population; set the maximum number of iterations T max , set the number of iterations t, t=1, 2, …, T max ; The iteration factor is E; According to the wavelet denoising parameter setting, the encoding individual H k And ranging laser signal information X n Perform simulation calculations to obtain the wavelet denoising parameter settings for the encoding individual H k The fitness of A k ; The specific steps of simulation calculation are: Using the formula The calculated wavelet denoising parameters are set to encode individual H k The fitness of A k ; Set the ranging laser signal information X n The length of time is D, X n (d) represents the ranging laser signal information at the dth moment, d = 1, 2, ..., D; U n (d) shows the use of wavelet denoising parameters to set the encoding individual H k The wavelet denoising parameter setting in encodes the simulated denoised ranging laser signal information after denoising at the dth moment; Using the formula Calculate the iteration factors E and F max is the maximum dynamic factor, F min is the minimum dynamic factor; when the iteration factor E>1, the wavelet denoising parameter setting encoding iteration population is updated globally; when the iteration factor E<1, the wavelet denoising parameter setting encoding iteration population is updated in the position with larger fitness; When the maximum number of iterations T is reached max When the output fitness is the largest, the wavelet denoising parameter setting corresponding to the encoding individual H k , which is the optimal wavelet denoising parameter setting encoding individual; according to the wavelet denoising parameter setting encoding in the optimal wavelet denoising parameter setting encoding individual, the ranging laser signal information X n Perform wavelet denoising.
6. The distance measurement method of laser medical equipment according to claim 5, characterized in that: The ranging error compensation model in step S3 includes a feature extraction layer, an error compensation factor matching layer and an error compensation factor output layer; The feature extraction layer is used to measure the external interference information G m Perform feature analysis to obtain the external interference information feature G of ranging m '; The error compensation factor matching layer is used to measure the external interference information characteristics G m 'Perform characteristic analysis and obtain the external interference error factor C m ; The error compensation factor output layer is used to output the external interference error factor C m ; The specific steps of training the error compensation factor matching layer include: Collecting several groups of error factor matching training samples, each group of error factor matching training samples contains a group of external interference information features and target values, and the target value is a corresponding interference error factor; combining several groups of error factor matching training samples to obtain an error compensation factor matching training set; The error compensation factor matching training set is input into the ranging error compensation model to train the error compensation factor matching layer with the target value as the target, and obtain the initial error compensation factor matching layer; the initial error compensation factor matching layer is model evaluated to obtain the initial error compensation factor matching layer model evaluation result; if the initial error compensation factor matching layer model evaluation result is passed, the initial error compensation factor matching layer is used as the error compensation factor matching layer in the ranging error compensation model; otherwise, the error compensation factor matching training set is used to continue the model training.
7. A distance measurement method for laser medical equipment according to claim 6, characterized in that: Based on the comprehensive external error interference factor C' and all the distances to be fitted L n The specific steps for data fitting include: Based on all the distances L to be fitted n Calculate the mean and get the mean of the fitting distance L'; use the formula The distance of the object to be measured is calculated, J represents the distance of the object to be measured, and P represents the basic error parameter.
8. A distance measurement system for laser medical equipment, characterized in that: The system uses a distance measurement method for laser medical equipment as described in any one of claims 1 to 7, including: The ranging scheme matching module is used to deploy the ranging scheme matching model, obtain the information of the object to be measured, the information of the object to be measured includes the starting point to be measured, the end point to be measured and the feature information of the object to be measured; the feature information of the object to be measured includes the image information of the object to be measured and the text information of the object to be measured; the ranging strategy of the object to be measured is obtained by analyzing the feature information of the object to be measured and the ranging scheme matching model; the ranging strategy of the object to be measured includes the initial detection threshold of the ranging of the object to be measured and the ranging accuracy standard of the object to be measured; the ranging accuracy standard of the object to be measured includes the number of times the ranging of the object to be measured is performed; the ranging scheme matching model is used to match ranging strategies of different accuracy according to different features of the object to be measured; the starting point to be measured, the end point to be measured, the initial detection threshold of the ranging of the object to be measured and the ranging accuracy standard of the object to be measured are combined to obtain the initial ranging scheme of the object to be measured; The ranging scheme execution module is used to deploy the ranging laser adaptive threshold model, and perform N distance measurements on the object to be measured based on the ranging accuracy standard of the object to be measured, where N is the specific number of times the distance measurement is performed on the object to be measured; the initial detection threshold of the distance measurement of the object to be measured is recorded as the distance measurement adjustment detection threshold Y1 of the object to be measured; for the nth distance measurement, n=1, 2, ..., N, the detection threshold Y is adjusted using the distance measurement of the object to be measured n The distance to be fitted is obtained by measuring the distance of the object to be measured using the distance measurement strategy of the object to be measured. n ; Get ranging laser signal information X n , based on the ranging laser signal information X n Analyze the ranging laser adaptive threshold model to obtain the distance adjustment detection threshold Y of the object to be measured n+1 ; In the ranging laser adaptive threshold model, the improved wavelet denoising method is used to denoise the laser signal in each ranging, and the denoised signal information and the ranging laser adaptive threshold adjustment function are used to analyze and change the detection threshold to improve the ranging accuracy; the ranging process is traversed N times until the distance to be fitted L1, the distance to be fitted L2, ..., the distance to be fitted L N ; The ranging error compensation module is used to deploy the ranging error compensation model and obtain the ranging external interference information G during the N ranging processes according to the preset time interval. m , m is the number of times the ranging external interference information is obtained at a preset time interval during the N ranging processes, m=1, 2, ..., M; according to the ranging external interference information G m The error analysis is performed with the ranging error compensation model to obtain the external interference error factor C m ; The ranging error compensation model is used to analyze the external interference information in the ranging process to reduce the error; for all external interference error factors C m Perform data fusion to obtain the comprehensive external error interference factor C'; based on the comprehensive external error interference factor C' and all the distances to be fitted L n Perform data fitting to obtain the distance of the object to be measured.
Citation Information
Patent Citations
Optimized-LM-algorithm-based laser radar waveform decomposition method
CN105093207A
Power transmission line laser ranging signal denoising algorithm based on lifting wavelet improved threshold
CN107765259A
Laser radar echo signal noise reduction method based on parameter optimization VMD
CN110865357A
Method and apparatus for improving signal to noise ratio using wavelet decomposition and frequency thresholding
EP1170698A2
Auto-ranging parametric spatial filtering
US10460429B1
