A method, device, electronic device and storage medium for obstacle avoidance navigation for the blind

Abstract

This application discloses a method, device, electronic device, and storage medium for obstacle avoidance navigation for the blind, relating to the field of computer technology. The method includes: performing target detection on the current environmental image of a preset environmental area to obtain the obstacle type of each obstacle; determining the obstacle weight factor of the corresponding obstacle based on the obstacle type of each obstacle; and determining the speed sensitivity coefficient of the corresponding obstacle based on the current obstacle data of each obstacle and the historical obstacle data of the corresponding obstacle; determining the noise interference coefficient of the preset environmental area at the current moment based on the current signal strength in the current voice data of the preset environmental area; and determining the obstacle avoidance risk index of the corresponding obstacle based on the noise interference coefficient, the current obstacle data of each obstacle, the obstacle weight factor, and the speed sensitivity coefficient; and performing graded obstacle avoidance navigation for the blind based on the obstacle avoidance risk index of each obstacle, which can effectively help the blind avoid dangers in complex environments.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, device, electronic device, and storage medium for obstacle avoidance navigation for the blind. Background Technology

[0002] Currently, blind people are mainly guided by guide dogs and guide canes for obstacle avoidance and navigation. However, the method of using guide dogs for obstacle avoidance and navigation has problems such as high training costs, limited numbers, long training cycles, and location restrictions. Furthermore, the method of using guide canes for obstacle avoidance and navigation can only perceive nearby obstacles through physical contact and cannot detect obstacles that are far away, suspended, or moving rapidly in advance. This makes it difficult for blind people to effectively avoid dangers in complex environments, which seriously limits their autonomy and safety when traveling. Summary of the Invention

[0003] This application provides a method, device, electronic device, and storage medium for obstacle avoidance navigation for the blind, which realizes the obstacle avoidance navigation function for the blind and solves the problem that blind people have difficulty effectively avoiding dangers in complex environments in the prior art.

[0004] In a first aspect, embodiments of this application provide a method for obstacle avoidance navigation for the blind, applied to an intelligent control device in a blind obstacle avoidance navigation system. The method includes: acquiring a current obstacle dataset, a current environmental image, and current voice data obtained at the current moment from a preset environmental area; the preset environmental area is an area centered on the blind person's location with a preset distance as its radius; performing target detection on the current environmental image to obtain the obstacle type of each obstacle; determining the obstacle weight factor of the corresponding obstacle based on the obstacle type of each obstacle; and determining the speed sensitivity coefficient of the corresponding obstacle based on the current obstacle data of each obstacle and the historical obstacle data of the corresponding obstacle; the current obstacle dataset includes the current obstacle data of each obstacle; determining the noise interference coefficient of the preset environmental area at the current moment based on the current signal strength in the current voice data; and determining the obstacle avoidance risk index of the corresponding obstacle based on the noise interference coefficient, the current obstacle data of each obstacle, the obstacle weight factor, and the speed sensitivity coefficient; and performing graded obstacle avoidance navigation for the blind person based on the obstacle avoidance risk index of each obstacle.

[0005] Secondly, embodiments of this application provide a blind obstacle avoidance navigation device, applied to an intelligent control device in a blind obstacle avoidance navigation system. The device includes: an acquisition module, used to acquire a current obstacle dataset, a current environmental image, and current voice data obtained at the current moment from a preset environmental area; the preset environmental area is an area centered on the blind person's location with a preset distance as its radius; a first determination module, used to perform target detection on the current environmental image to obtain the obstacle type of each obstacle, determine the obstacle weight factor of the corresponding obstacle based on the obstacle type of each obstacle, and determine the speed sensitivity coefficient of the corresponding obstacle based on the current obstacle data of each obstacle and the historical obstacle data of the corresponding obstacle; the current obstacle dataset includes the current obstacle data of each obstacle; a second determination module, used to determine the noise interference coefficient of the preset environmental area at the current moment based on the current signal strength in the current voice data, and determine the obstacle avoidance risk index of the corresponding obstacle based on the noise interference coefficient, the current obstacle data of each obstacle, the obstacle weight factor, and the speed sensitivity coefficient; and an obstacle avoidance module, used to perform graded obstacle avoidance navigation for the blind person based on the obstacle avoidance risk index of each obstacle.

[0006] Thirdly, embodiments of this application provide an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to execute the obstacle avoidance navigation method for the blind according to any embodiment of this application.

[0007] Fourthly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the obstacle avoidance navigation method for the blind as described in any embodiment of this application.

[0008] In this embodiment, the current obstacle dataset, current environmental image, and current voice data obtained at the current moment can be acquired from a preset environmental area. The preset environmental area is an area centered on the location of the blind person with a preset distance as its radius. Then, target detection is performed on the current environmental image to obtain the obstacle type of each obstacle. Based on the obstacle type of each obstacle, the obstacle weight factor of the corresponding obstacle is determined. Then, based on the current obstacle data and the historical obstacle data of the corresponding obstacle, the speed sensitivity coefficient of the corresponding obstacle is determined. Based on the current signal strength in the current voice data, the noise interference coefficient of the preset environmental area at the current moment is determined. This provides an accurate data foundation for the subsequent calculation of the obstacle avoidance risk index. Afterward, based on the noise interference coefficient, the current obstacle data, obstacle weight factor, and speed sensitivity coefficient of each obstacle, the obstacle avoidance risk index of the corresponding obstacle is determined. This can comprehensively quantify the collision risk level of the obstacle to the blind person from both dynamic (i.e., the relative motion between the obstacle and the blind person) and static (i.e., the inherent danger of the obstacle itself and environmental noise) perspectives, thereby improving the accuracy of the obstacle avoidance risk index determination. Based on the obstacle avoidance risk index of each obstacle, the blind person can be guided to perform graded obstacle avoidance navigation. In the above technical solution, based on obstacle data, environmental images, and voice data within a preset environmental area, the obstacle avoidance risk index of each obstacle within the preset environmental area is determined, and graded obstacle avoidance navigation is performed. This allows for the early detection of obstacles that are far away, suspended, or moving rapidly, and timely prompts for blind people to avoid obstacles. This achieves automated obstacle avoidance navigation for blind people, effectively avoiding the problems of high breeding costs, limited numbers, long training cycles, and location restrictions associated with guide dogs, as well as the problem that guide canes can only detect obstacles at close range. In this way, it can effectively help blind people avoid dangers in complex environments, providing more reliable safety guarantees for blind people's travel, thereby greatly improving the autonomy and safety of blind people's travel. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a flowchart illustrating a method for obstacle avoidance navigation for the blind provided in an embodiment of this application;

[0011] Figure 2 This is another flowchart illustrating the obstacle avoidance navigation method for the blind provided in this application embodiment;

[0012] Figure 3This is a schematic diagram of a blind obstacle avoidance navigation device provided in an embodiment of this application;

[0013] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0015] It should be noted that the terms "first," "second," "target," and "original," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in sequences other than those illustrated or described herein. Furthermore, the terms "comprising," "having," and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0016] Figure 1 This is a flowchart illustrating a method for obstacle avoidance navigation for the blind provided in this application embodiment. This embodiment can be applied to scenarios requiring obstacle avoidance navigation for the blind. The obstacle avoidance navigation method for the blind provided in this embodiment can be executed by the obstacle avoidance navigation device provided in this application embodiment. This device can be implemented through software and / or hardware. In a specific embodiment, the obstacle avoidance navigation device can be integrated into an electronic device, which is an intelligent control device in the obstacle avoidance navigation system for the blind. For example, the electronic device can be a portable device such as a smartphone, smartwatch, or smart bracelet. The executing entity of this method can be an electronic device (i.e., the intelligent control device in the obstacle avoidance navigation system for the blind).

[0017] In one specific embodiment, the obstacle avoidance navigation system for the blind may include an intelligent control device, smart glasses, and headphones; wherein, the intelligent control device can wirelessly communicate with the smart glasses and headphones, and is responsible for receiving and fusing the collected data transmitted by the smart glasses and headphones in real time, such as data collected by lidar devices, ultrasonic sensors, and cameras, to quickly identify obstacles, calculate their precise positions, and predict the movement trajectory of moving obstacles.

[0018] The smart glasses feature a lightweight design, with a weight controlled within a preset range (e.g., 50 grams). They utilize an ergonomic frame to ensure comfort during extended wear. Furthermore, the smart glasses are equipped with distance detection devices (such as lidar and ultrasonic sensors) and cameras. For instance, a high-precision lidar device is installed at the front of the frame to detect the relative distance, azimuth, and speed of obstacles within a long distance range (e.g., 5 meters to 50 meters). Miniature ultrasonic sensors are installed on the temples to detect the relative distance, azimuth, and speed of obstacles within a short distance range (e.g., 0.1 meters to 5 meters), compensating for the lidar device's blind spots at close range. A low-power, ultra-wide-angle camera is installed above the frame to identify obstacle types (e.g., pedestrians, vehicles, steps, or utility poles).

[0019] The earphones are equipped with a microphone and use advanced bone conduction technology to transmit sound directly to the inner ear through the skull. This ensures clear information transmission without affecting the blind person's perception of surrounding sounds, thus guaranteeing travel safety.

[0020] See Figure 1 The obstacle avoidance navigation method for the blind in this embodiment includes, but is not limited to, the following steps:

[0021] S110. Obtain the current obstacle dataset, current environment image, and current voice data obtained from the preset environmental area at the current moment.

[0022] The preset environment area is a region centered on the location of the blind person and with a preset distance as its radius. The preset distance is a pre-set distance used to characterize obstacles within the preset distance of the blind person's location that can be detected in advance. Here, obstacles are all objects that hinder the blind person's movement.

[0023] The current obstacle dataset contains relevant information about all obstacles within the preset environmental area at the current moment, including current obstacle data for all obstacles within the preset environmental area at the current moment. Optionally, the current obstacle data may include current relative speed, current relative distance, and current relative azimuth. The current relative speed is the relative speed between the obstacle and the blind person at the current moment; the current relative distance is the relative distance between the obstacle and the blind person at the current moment; and the current relative azimuth is the relative azimuth between the obstacle and the blind person at the current moment.

[0024] The current environment image is an image used to visually represent the scene within a preset environmental area at the current moment. The current speech data is the speech signal within the preset environmental area at the current moment; optionally, the current speech data may include the current signal frequency and the current signal strength.

[0025] Specifically, at the current moment, an obstacle dataset obtained by using a distance detection device to collect data from a preset environmental area can be acquired, namely the current obstacle dataset. That is, an obstacle dataset obtained by using an ultrasonic sensor to collect data from a preset environmental area can be acquired, denoted as the first obstacle dataset, and an obstacle dataset obtained by using a lidar device to collect data from a preset environmental area can be acquired, denoted as the second obstacle dataset. Then, the first obstacle dataset and the second obstacle dataset are fused to obtain the current obstacle dataset, and the current obstacle data and the timestamp corresponding to the current obstacle data (i.e., the current moment) in the current obstacle dataset are saved.

[0026] Specifically, since lidar devices have detection blind spots at close range, while ultrasonic sensors do not, obstacle data within a preset close range is obtained from a first obstacle dataset to form a first intermediate obstacle dataset, and obstacle data within a preset long range is obtained from a second obstacle dataset to form a second intermediate obstacle dataset. The first and second intermediate obstacle datasets are then concatenated to obtain the current obstacle dataset. Alternatively, a Karman filter can be used to fuse the first and second obstacle datasets to obtain the current obstacle dataset, which can improve the accuracy of obtaining the current obstacle dataset. Here, the preset close range is a distance range pre-set based on the lidar device's detection blind spot, and the preset long range is a distance range pre-set based on the lidar device's detection blind spot, but not within the lidar device's detection blind spot.

[0027] Then, it can acquire the environmental image obtained by using the camera to capture the preset environmental area, i.e., the current environmental image, and acquire the voice data obtained by using the microphone to capture the preset environmental area, i.e., the current voice data.

[0028] S120. Perform target detection on the current environment image to obtain the obstacle type of each obstacle. Determine the obstacle weight factor of the corresponding obstacle based on the obstacle type of each obstacle. Determine the speed sensitivity coefficient of the corresponding obstacle based on the current obstacle data of each obstacle and the historical obstacle data of the corresponding obstacle.

[0029] The obstacle weight factor is used to quantify the collision risk to blind people caused by the inherent danger of the obstacle itself. Different obstacle types correspond to different obstacle weight factors. The greater the inherent danger of the obstacle itself, the greater the obstacle weight factor of the corresponding type, and the higher the collision risk. For example, the obstacle weight factor of vehicle type is greater than that of pedestrian type obstacle.

[0030] Historical obstacle data refers to the obstacle data within the first historical time period, which is a pre-set time range used to determine whether the obstacle is in motion.

[0031] The speed sensitivity coefficient is used to amplify or reduce the collision risk posed to a blind person by the relative motion between the obstacle and the blind person. Different speed sensitivity coefficients are corresponding to obstacles with different dynamic characteristics (i.e., moving and stationary states). Since moving obstacles require more avoidance from the blind person than stationary obstacles, a larger speed sensitivity coefficient is set for moving obstacles to amplify the collision risk posed to the blind person; and a smaller speed sensitivity coefficient is set for stationary obstacles to reduce the collision risk posed to the blind person.

[0032] Specifically, target detection algorithms can be used to detect targets in the current environment image to obtain the obstacle types of each obstacle. For example, the current environment image can be input into a pre-trained target detection model to obtain the obstacle types of each obstacle. Then, obstacle weight factors of the corresponding obstacles can be determined based on the obstacle types of each obstacle. For example, for the current obstacle among the obstacles, a preset type weight relationship can be queried based on the obstacle type of the current obstacle to obtain the obstacle weight factor corresponding to the current obstacle. The preset type weight relationship is a pre-set mapping relationship used to store the correspondence between obstacle types and obstacle weight factors.

[0033] Then, the speed sensitivity coefficient of the corresponding obstacle can be determined based on the current obstacle data and the historical obstacle data of the corresponding obstacle. That is, for the current obstacle in each obstacle, the current obstacle data of the current obstacle can be obtained from the current obstacle dataset, and the obstacle data of the current obstacle in the first historical time period can be obtained to obtain the historical obstacle data. Then, based on the current obstacle data and the historical obstacle data of the current obstacle, it can be determined whether the current obstacle is in motion or in a stationary state.

[0034] Specifically, the geographical location of the blind person can be detected using the Global Positioning System (GPS) installed within the electronic device. Then, based on the blind person's geographical location, the current relative distance and azimuth angle from the current obstacle data, the geographical location of the obstacle at the current moment is determined. Following the same calculation steps, the geographical location of the obstacle at each moment within a first historical time period is determined. The distance between the current geographical location of the obstacle and the geographical locations at each moment within the first historical time period is then calculated. If the distance between the current geographical location of the obstacle and the geographical locations at each moment within the first historical time period is less than a preset error distance, the obstacle is determined to be stationary, and its velocity sensitivity coefficient is set to the preset velocity sensitivity coefficient for a stationary state. Otherwise, the obstacle is determined to be moving, and its velocity sensitivity coefficient is set to the preset velocity sensitivity coefficient for a moving state. The preset error distance is a pre-set distance used to characterize the maximum positional change caused by the detection error of the distance detection device.

[0035] S130. Determine the noise interference coefficient of the preset environment area at the current moment based on the current signal strength in the current voice data, and determine the obstacle avoidance risk index of the corresponding obstacle based on the noise interference coefficient, the current obstacle data of each obstacle, the obstacle weight factor and the speed sensitivity coefficient.

[0036] The noise interference coefficient is used to quantify the degree to which environmental noise interferes with the effectiveness of obstacle avoidance voice prompts. The obstacle avoidance risk index is used to quantify the overall collision risk posed by obstacles to blind people; the higher the obstacle avoidance risk index, the higher the corresponding collision risk.

[0037] Specifically, the noise interference coefficient of a preset environmental area at the current moment can be determined based on the current signal strength and a preset strength threshold. The specific calculation formula is as follows: Among them, C noise_i Let be the noise interference coefficient at time i, in dB. iLet be the signal strength at time i, and b be the preset strength threshold, which is a fixed value used to characterize the critical value at which environmental noise interferes with the obstacle avoidance voice prompt effect.

[0038] Then, the obstacle avoidance risk index of the corresponding obstacle can be determined based on the noise interference coefficient, the current obstacle data, obstacle weighting factor and speed sensitivity coefficient of each obstacle. For example, for the current obstacle among the obstacles, the dynamic collision risk level caused by the relative motion between the current obstacle and the blind person can be determined based on the current obstacle data and speed sensitivity coefficient of the current obstacle. The inherent danger of the current obstacle and the static collision risk level caused by environmental noise to the blind person can be determined based on the noise interference coefficient and the obstacle weighting factor of the current obstacle. Then, the comprehensive collision risk level caused by the current obstacle to the blind person can be determined based on the dynamic collision risk level and the static collision risk level, that is, the obstacle avoidance risk index corresponding to the current obstacle.

[0039] S140. Classified obstacle avoidance navigation for blind people based on the obstacle avoidance risk index of each obstacle.

[0040] Specifically, obstacle avoidance navigation for the blind can be tiered based on the obstacle avoidance risk index of each obstacle. For example, a preset risk warning relationship can be queried based on the obstacle avoidance risk index of each obstacle to obtain the obstacle avoidance warning mode for the corresponding obstacle. The preset risk warning relationship is a pre-set correspondence used to store the correspondence between the value range of the obstacle avoidance risk index and the obstacle avoidance warning mode. The obstacle avoidance warning mode can include a no-warning mode, a voice warning mode, and a voice warning and vibration warning mode, etc. Then, obstacles whose obstacle avoidance warning mode is not a no-warning mode are identified and recorded as target obstacles. Based on the obstacle avoidance warning mode of each target obstacle, the current obstacle data, and the obstacle type, the corresponding obstacle avoidance warning text for the target obstacle is generated. The obstacle avoidance warning text is used to instruct the blind person to avoid the obstacle, and the corresponding voice is generated. Then, according to the obstacle avoidance warning mode, the corresponding voice of the obstacle avoidance warning text for the target obstacle is played to the blind person through headphones to guide the blind person to avoid the obstacle. For obstacles whose obstacle avoidance warning mode is a no-warning mode, it is not necessary to generate and play the corresponding obstacle avoidance warning text. This achieves tiered obstacle avoidance navigation for the blind.

[0041] The technical solution of this application embodiment can acquire the current obstacle dataset, current environment image, and current voice data obtained from a preset environmental area at the current moment. The preset environmental area is an area centered on the location of the blind person with a preset distance as its radius. Then, target detection is performed on the current environmental image to obtain the obstacle type of each obstacle. Based on the obstacle type of each obstacle, the obstacle weight factor of the corresponding obstacle is determined. Then, based on the current obstacle data and the historical obstacle data of the corresponding obstacle, the speed sensitivity coefficient of the corresponding obstacle is determined. Based on the current signal strength in the current voice data, the noise interference coefficient of the preset environmental area at the current moment is determined. This provides an accurate data foundation for the subsequent calculation of the obstacle avoidance risk index. Afterwards, based on the noise interference coefficient, the current obstacle data of each obstacle, the obstacle weight factor, and the speed sensitivity coefficient, the obstacle avoidance risk index of the corresponding obstacle is determined. This can comprehensively quantify the collision risk level of the obstacle to the blind person from both a dynamic perspective (i.e., the relative motion between the obstacle and the blind person) and a static perspective (i.e., the inherent danger of the obstacle itself and environmental noise), thereby improving the accuracy of the obstacle avoidance risk index determination. Based on the obstacle avoidance risk index of each obstacle, the blind person can be guided to perform graded obstacle avoidance navigation. In the above technical solution, based on obstacle data, environmental images, and voice data within a preset environmental area, the obstacle avoidance risk index of each obstacle within the preset environmental area is determined, and graded obstacle avoidance navigation is performed. This allows for the early detection of obstacles that are far away, suspended, or moving rapidly, and timely prompts for blind people to avoid obstacles. This achieves automated obstacle avoidance navigation for blind people, effectively avoiding the problems of high breeding costs, limited numbers, long training cycles, and location restrictions associated with guide dogs, as well as the problem that guide canes can only detect obstacles at close range. In this way, it can effectively help blind people avoid dangers in complex environments, providing more reliable safety guarantees for blind people's travel, thereby greatly improving the autonomy and safety of blind people's travel.

[0042] The following further describes a method for obstacle avoidance navigation for the blind provided by an embodiment of this application. Figure 2 This is another schematic flowchart of the obstacle avoidance navigation method for the blind provided in this application. This application's embodiment is an optimization based on the above embodiments. See also... Figure 2 The method in this embodiment includes, but is not limited to, the following steps:

[0043] S210. Obtain the current obstacle dataset, current environment image, and current voice data obtained from the preset environmental area at the current moment.

[0044] S220. Perform target detection on the current environment image to obtain the obstacle type of each obstacle.

[0045] S230. Based on the current warning distance and the obstacle type of each obstacle, query the first preset mapping relationship to obtain the obstacle weight factor of the corresponding obstacle.

[0046] The first preset mapping relationship includes the correspondence between warning distance, obstacle type, and obstacle weight factor. The warning distance is the minimum safe distance between the blind person and the obstacle, and it is updated every preset time interval. When the relative distance between the blind person and the obstacle is greater than or equal to the minimum safe distance, there is no need to prompt the blind person to avoid the obstacle, i.e., no obstacle avoidance prompt text needs to be generated and broadcast. When the relative distance between the blind person and the obstacle is less than the minimum safe distance, there is a need to prompt the blind person to avoid the obstacle, i.e., the obstacle avoidance prompt text needs to be generated and broadcast. The preset time interval is the pre-set update cycle of the warning distance. The current warning distance is the latest warning distance, i.e., the most recently updated warning distance.

[0047] Optionally, the steps for determining the current warning distance are as follows, including Sa1-Sa3:

[0048] Sa1. Obtain the average reaction time of blind people over a historical period to get the historical average reaction time.

[0049] Among them, reaction time is the time difference between the moment when the blind person makes an obstacle avoidance reaction and the moment when the voice broadcast of the corresponding obstacle avoidance prompt text ends. Historical average reaction time is the average reaction time of the blind person within a historical period. Here, historical period is the time range corresponding to the previous update cycle of the warning distance. For example, if the time range corresponding to the previous update cycle of the warning distance is [8:30, 8:40], then the historical period is [8:30, 8:40]. The preset period at this time is 10 minutes.

[0050] Specifically, after each obstacle avoidance voice prompt to the blind person, the time when the voice prompt ends and the time when the blind person makes an obstacle avoidance response are recorded. The difference between the time when the blind person makes an obstacle avoidance response and the time when the voice prompt ends is calculated to obtain the blind person's reaction time for that moment. Thus, the reaction time of the blind person at every moment in the historical time period can be obtained, and the average of these reaction times can be calculated to obtain the historical average reaction time.

[0051] Sa2. Based on the historical average reaction time, the preset baseline reaction time, and the preset compensation scaling factor, determine the candidate distance compensation ratio, and based on the candidate distance compensation ratio and the preset maximum compensation coefficient, determine the target distance compensation ratio.

[0052] The preset baseline reaction time is a pre-set average reaction time used to characterize the maximum average reaction time of a healthy adult in a conscious state. When the average reaction time of a blind person is less than the preset baseline reaction time, it indicates that the blind person reacts quickly and does not need to compensate for the warning distance, and the warning distance can be appropriately reduced; otherwise, it indicates that the blind person reacts slowly and needs to compensate for the warning distance, that is, the warning distance needs to be increased.

[0053] A preset compensation scaling factor is used to amplify or reduce the compensation intensity of the warning distance, mapping the average reaction delay to the corresponding distance compensation ratio. This controls the rate and range of distance compensation, preventing over-compensation of the warning distance and avoiding drastic changes in the warning distance due to small changes in the average reaction time. For example, the preset compensation scaling factor can be 0.5. The average reaction delay is the difference between the average reaction time and the preset baseline reaction time.

[0054] The candidate distance compensation ratio is a distance compensation ratio determined based on the average reaction delay of blind people. The preset maximum compensation coefficient is a constraint on the pre-set distance compensation ratio to avoid false alarms and confusion caused by a sudden increase in warning distance due to abnormal reaction time (such as a blind person falling and having a longer reaction time). For example, the preset maximum compensation coefficient can be 0.2. The target distance compensation ratio is the distance compensation ratio obtained under the constraint of the preset maximum compensation coefficient.

[0055] Specifically, the difference between the historical average reaction time and the preset baseline reaction time can be calculated to obtain the historical average reaction delay of the blind person. The historical average reaction delay may be positive or negative. Then, the product of the historical average reaction delay and the preset compensation scaling factor is calculated to obtain the candidate distance compensation ratio. Finally, the product of the candidate distance compensation ratio and the preset maximum compensation coefficient is calculated to obtain the target distance compensation ratio.

[0056] Sa3. Calculate the product of the target distance compensation ratio and the preset baseline warning distance to obtain the distance compensation amount, and calculate the sum of the preset baseline warning distance and the distance compensation amount to obtain the current warning distance.

[0057] The preset baseline warning distance is the pre-set default warning distance. The distance compensation amount is determined based on the historical average reaction time.

[0058] In this embodiment, the warning distance can be dynamically adjusted based on the average reaction time of blind people in the recent historical period. This can improve computational efficiency, reduce implementation complexity, and thus improve the accuracy and efficiency of determining the current warning distance. This makes the current warning distance highly adaptable to the average reaction time of blind people in the recent period, thereby reserving sufficient obstacle avoidance reaction time for blind people and improving the safety of blind people.

[0059] S240. Determine the speed sensitivity coefficient of the corresponding obstacle based on the current obstacle data and the historical obstacle data of the corresponding obstacle.

[0060] S250. Determine the noise interference coefficient of the preset environmental area based on the current signal strength in the current voice data, and determine the obstacle avoidance risk index of the corresponding obstacle based on the noise interference coefficient, the current obstacle data of each obstacle, the obstacle weight factor and the speed sensitivity coefficient.

[0061] Specifically, for the current obstacle among all obstacles, the target approach rate can be obtained by multiplying the speed sensitivity coefficient of the current obstacle by the absolute value of its current relative velocity. This amplifies the approach rate of moving obstacles and reduces the approach rate of stationary obstacles. Since relative velocity is a directional vector, and the direction in which the obstacle approaches the blind person is set as the positive direction and the direction away from the blind person is set as the negative direction, the target approach rate is determined here using the absolute value of the current relative velocity. The target approach rate is a value obtained after amplifying or reducing the current relative velocity according to the dynamic characteristics of the obstacle. Next, the ratio between the target approach rate and the current relative distance of the current obstacle is calculated. The process involves several steps: First, a dynamic risk index is obtained, which quantifies the collision risk posed by the relative motion between an obstacle and a blind person. The dynamic risk index for an obstacle in motion with the same relative distance and speed is higher than that for a stationary obstacle, ensuring that moving obstacles require greater avoidance from the blind. Then, the product of the noise interference coefficient and the obstacle weight factor of the current obstacle is calculated to obtain a static risk index. This static risk index quantifies the collision risk posed by the inherent hazard of the obstacle and environmental noise. The sum of the dynamic and static risk indices is then calculated to obtain the obstacle avoidance risk index for the current obstacle. This approach comprehensively considers the collision risk posed by obstacles from both dynamic and static perspectives, improves computational efficiency, reduces implementation complexity, and thus enhances the accuracy and efficiency of obstacle avoidance index determination. This results in higher obstacle avoidance risk indices for moving obstacles with high inherent hazard within a high-noise environment, providing an accurate data foundation for subsequent graded obstacle avoidance navigation.

[0062] S260. Determine the obstacle avoidance risk level of the corresponding obstacle based on the obstacle avoidance risk index of each obstacle, and select the prompt semantic template of the corresponding obstacle from at least two preset semantic templates based on the obstacle avoidance risk level.

[0063] The obstacle avoidance risk level is the level corresponding to the degree of collision risk posed by obstacles to a blind person; the higher the obstacle avoidance risk index, the higher the corresponding obstacle avoidance risk level and the higher the degree of collision risk. The preset semantic template is a pre-set text template with a fixed semantic structure and clear information dimensions, and the correspondence between the obstacle avoidance risk level and the preset semantic template is pre-set. For example, the preset semantic template may include an emergency semantic template and a standard semantic template. The emergency semantic template can be [obstacle avoidance suggestion], and the standard semantic template can be [direction] [distance] meters [obstacle] [obstacle avoidance suggestion].

[0064] Specifically, for the current obstacle among all obstacles, the obstacle avoidance risk index of the current obstacle can be compared with the obstacle avoidance risk index range corresponding to the obstacle avoidance risk level to determine which obstacle avoidance risk index range the current obstacle's obstacle avoidance risk index falls into, thereby determining the obstacle avoidance risk level of the current obstacle; then, based on the obstacle avoidance risk level of the current obstacle, the correspondence between the obstacle avoidance risk level and the preset semantic template is queried to obtain the preset semantic template corresponding to the current obstacle, and the preset semantic template is obtained from at least two preset semantic templates to obtain the prompt semantic template of the current obstacle.

[0065] S270. Generate obstacle avoidance prompt text for each obstacle based on the current obstacle data, obstacle type, and prompt semantic template, and provide obstacle avoidance navigation for the blind based on the obstacle avoidance prompt text for each obstacle.

[0066] Specifically, for the current obstacle among all obstacles, obstacle avoidance suggestions can be determined based on the current relative distance, current relative speed, current relative azimuth angle and obstacle type of the current obstacle. Then, obstacle avoidance prompt text for the current obstacle can be generated according to the prompt semantic template based on the current relative distance, current relative speed, current relative azimuth angle and the required data in the obstacle avoidance suggestions. Finally, obstacle avoidance navigation is performed for blind people based on the obstacle avoidance prompt text for each obstacle.

[0067] Furthermore, obstacle avoidance navigation is provided for blind users based on obstacle avoidance prompts for each obstacle, including Sb1-Sb3:

[0068] Sb1. Determine the obstacle avoidance prompt voice intensity based on the current signal frequency and the preset noise main frequency band.

[0069] The preset noise dominant frequency band is a specific frequency range that is pre-set to have a major impact on obstacle avoidance voice prompts. The obstacle avoidance prompt voice intensity is the voice intensity of the obstacle avoidance prompt text being broadcast.

[0070] Specifically, in one implementation, if the current signal frequency is within a preset noise main frequency band, the obstacle avoidance prompt voice intensity is determined to be a first voice prompt intensity; if the current signal frequency is lower than the preset noise main frequency band, the obstacle avoidance prompt voice intensity is determined to be a second voice prompt intensity; if the current signal frequency is higher than the preset noise main frequency band, the obstacle avoidance prompt voice intensity is determined to be a third voice prompt intensity. The first, second, and third voice prompt intensities are preset voice intensities, and the first voice prompt intensity is less than the second voice prompt intensity, and the second voice prompt intensity is less than the third voice prompt intensity.

[0071] In another implementation, when the current signal frequency is within the preset noise main frequency band, the obstacle avoidance prompt voice intensity is determined to be the sum of the preset baseline prompt voice intensity and the preset intensity increase. When the current signal frequency is outside the preset noise main frequency band, the obstacle avoidance prompt voice intensity is determined to be the preset baseline prompt voice intensity. By specifically enhancing the voice intensity of the obstacle avoidance prompt within the preset noise main frequency band, the masking effect of noise on the obstacle avoidance voice prompt within the preset noise main frequency band can be broken, thereby achieving noise countermeasure function and enabling blind people to clearly distinguish the voice content of the obstacle avoidance prompt. The preset baseline prompt voice intensity is the default voice intensity pre-set for the obstacle avoidance prompt text; the preset intensity increase is the pre-set voice intensity increase.

[0072] Sb2. Based on the obstacle avoidance risk level of each obstacle, query the second preset mapping relationship to obtain the obstacle avoidance prompt speed of the corresponding obstacle, and determine the obstacle avoidance prompt order based on the obstacle avoidance risk level of each obstacle.

[0073] The second preset mapping relationship includes the correspondence between obstacle avoidance risk levels and obstacle avoidance prompt speech rate. The obstacle avoidance prompt speech rate is the speech rate at which the obstacle avoidance prompt text is read, and the speech rate of the obstacle avoidance prompt meets the clarity requirements.

[0074] The obstacle avoidance prompt order is the order in which the obstacle avoidance prompt text is broadcast. That is, the order of the obstacle avoidance prompts for each obstacle is determined based on the order of the obstacle avoidance risk level. Specifically, the higher the obstacle avoidance risk level, the earlier the obstacle avoidance prompt is broadcast; the lower the obstacle avoidance risk level, the later the obstacle avoidance prompt is broadcast, so that the obstacle avoidance prompt text corresponding to the obstacle with the high obstacle avoidance risk level is broadcast first.

[0075] It should be noted that for two obstacles with the same obstacle avoidance risk level, the higher the obstacle avoidance risk index, the earlier the obstacle avoidance prompt will appear; if the obstacle avoidance risk indices of the two obstacles are equal, the order of the obstacle avoidance prompts can be randomized.

[0076] Sb3. Based on the obstacle avoidance prompts, the intensity of the obstacle avoidance prompts, and the speech rate of the obstacle avoidance prompts for each obstacle, provide obstacle avoidance voice prompts to the blind person according to the obstacle avoidance prompt text for the corresponding obstacle.

[0077] In this embodiment, the obstacle avoidance prompt voice intensity is determined based on the current signal frequency and the preset noise main frequency band, which can improve the clarity of the obstacle avoidance voice prompt and reduce the impact of environmental noise on the obstacle avoidance voice prompt. Then, the speech rate of the obstacle avoidance prompt for each obstacle is determined, and the speech rate of the obstacle avoidance prompt can be adaptively adjusted based on the obstacle avoidance risk level, so that blind people can hear the obstacle avoidance prompt text in time. Then, the order of obstacle avoidance prompts is determined, and the obstacle avoidance prompt text of each obstacle is broadcast in the order of obstacle avoidance prompts. The obstacle avoidance prompt text corresponding to obstacles with high obstacle avoidance risk level can be broadcast first, thereby improving the accuracy and efficiency of graded obstacle avoidance navigation, thereby improving the safety of blind people traveling.

[0078] The technical solution of this application embodiment can acquire the current obstacle dataset, current environment image, and current voice data obtained from a preset environmental area at the current moment, and perform target detection on the current environment image to obtain the obstacle type of each obstacle. Next, based on the current warning distance and the obstacle type of each obstacle, a first preset mapping relationship is queried to obtain the obstacle weight factor of the corresponding obstacle. By looking up the table, the efficiency and accuracy of determining the obstacle weight factor can be improved. Then, based on the current obstacle data of each obstacle and the historical obstacle data of the corresponding obstacle, the speed sensitivity coefficient of the corresponding obstacle is determined, and based on the current voice data... Signal strength determines the noise interference coefficient of a preset environmental area. Based on the noise interference coefficient, current obstacle data, obstacle weighting factors, and velocity sensitivity coefficients of each obstacle, the obstacle avoidance risk index of the corresponding obstacle is determined. This allows for a comprehensive quantification of the collision risk posed by obstacles to the blind person from both dynamic (i.e., the relative motion between the obstacle and the blind person) and static (i.e., the inherent danger of the obstacle itself and environmental noise) perspectives, thereby improving the accuracy of the obstacle avoidance risk index determination. Then, based on the obstacle avoidance risk index of each obstacle, the obstacle avoidance risk level of the corresponding obstacle is determined. This allows for the classification of obstacle avoidance risk levels and the determination of at least two risk levels. The system selects a corresponding obstacle's semantic template from a preset semantic template. This template can be dynamically adjusted based on the obstacle avoidance risk level, providing an accurate data foundation for generating subsequent obstacle avoidance warning texts. Then, based on the current obstacle data, obstacle type, and semantic template, the system generates the corresponding obstacle avoidance warning text. This ensures that the warning text only includes the text required for the corresponding obstacle avoidance risk level, avoiding redundant text generation. It dynamically generates the most concise, relevant, and urgent obstacle avoidance warning text, allowing the generation rate to adapt to the corresponding obstacle avoidance risk level. Finally, it enables obstacle avoidance guidance for blind individuals based on the obstacle avoidance warning texts for each obstacle. Navigation technology allows obstacle avoidance to adapt its response rate to the corresponding obstacle risk level, thereby improving its accuracy and efficiency. Furthermore, it can detect distant, suspended, or fast-moving obstacles in advance and promptly alert blind users to avoid them, automating obstacle avoidance navigation for the blind. This effectively avoids the problems associated with guide dogs (high breeding costs, limited numbers, long training periods, and location restrictions) and guide canes (which can only detect nearby obstacles). Consequently, it can effectively help blind people avoid danger in complex environments, providing more reliable safety guarantees for their travels and greatly improving their autonomy and safety.

[0079] Figure 3 This is a schematic diagram of a blind obstacle avoidance navigation device provided in an embodiment of this application, referring to... Figure 3 The obstacle avoidance navigation device for the blind may include:

[0080] The acquisition module 310 is used to acquire the current obstacle dataset, current environment image and current voice data obtained at the current moment from the preset environmental area; the preset environmental area is an area centered on the location of the blind person with a preset distance as the radius;

[0081] The first determining module 320 is used to perform target detection on the current environment image, obtain the obstacle type of each obstacle, determine the obstacle weight factor of the corresponding obstacle based on the obstacle type of each obstacle, and determine the speed sensitivity coefficient of the corresponding obstacle based on the current obstacle data of each obstacle and the historical obstacle data of the corresponding obstacle. The current obstacle dataset includes the current obstacle data of each obstacle.

[0082] The second determining module 330 is used to determine the noise interference coefficient of the preset environmental area at the current moment based on the current signal strength in the current voice data, and to determine the obstacle avoidance risk index of the corresponding obstacle based on the noise interference coefficient, the current obstacle data of each obstacle, the obstacle weight factor and the speed sensitivity coefficient.

[0083] The obstacle avoidance module 340 is used to provide graded obstacle avoidance navigation for blind people based on the obstacle avoidance risk index of each obstacle.

[0084] In one embodiment, the current obstacle data includes the current relative speed and the current relative distance. The second determining module 330 determines the obstacle avoidance risk index of the corresponding obstacle based on the noise interference coefficient, the current obstacle data of each obstacle, the obstacle weight factor, and the speed sensitivity coefficient. This includes: for the current obstacle among all obstacles, calculating the product of the speed sensitivity coefficient of the current obstacle and the absolute value of the current relative speed of the current obstacle to obtain the target approach rate, and calculating the ratio between the target approach rate and the current relative distance of the current obstacle to obtain the dynamic risk index; calculating the product between the noise interference coefficient and the obstacle weight factor of the current obstacle to obtain the static risk index, and calculating the sum of the dynamic risk index and the static risk index to obtain the obstacle avoidance risk index of the current obstacle.

[0085] In one embodiment, the first determining module 320 determines the obstacle weight factor of the corresponding obstacle based on the obstacle type of each obstacle, including: querying a first preset mapping relationship based on the current warning distance and the obstacle type of each obstacle to obtain the obstacle weight factor of the corresponding obstacle; the first preset mapping relationship includes the correspondence between the warning distance, the obstacle type and the obstacle weight factor.

[0086] In one embodiment, the steps for determining the current warning distance in the first determining module 320 are as follows: obtaining the average reaction time of the blind person over a historical period to obtain the historical average reaction time; determining the candidate distance compensation ratio based on the historical average reaction time, the preset baseline reaction time, and the preset compensation scaling factor, and determining the target distance compensation ratio based on the candidate distance compensation ratio and the preset maximum compensation coefficient; calculating the product of the target distance compensation ratio and the preset baseline warning distance to obtain the distance compensation amount, and calculating the sum of the preset baseline warning distance and the distance compensation amount to obtain the current warning distance.

[0087] In one embodiment, the obstacle avoidance module 340 is specifically used to: determine the obstacle avoidance risk level of the corresponding obstacle based on the obstacle avoidance risk index of each obstacle, and select the prompt semantic template of the corresponding obstacle from at least two preset semantic templates based on the obstacle avoidance risk level; generate the obstacle avoidance prompt text of the corresponding obstacle based on the current obstacle data, obstacle type and prompt semantic template of each obstacle, and provide obstacle avoidance navigation for the blind based on the obstacle avoidance prompt text of each obstacle.

[0088] In one embodiment, the current voice data also includes the current signal frequency. The obstacle avoidance module 340 provides obstacle avoidance navigation for the blind person based on the obstacle avoidance prompt text for each obstacle, including: determining the intensity of the obstacle avoidance prompt voice based on the current signal frequency and a preset noise main frequency band; querying a second preset mapping relationship based on the obstacle avoidance risk level of each obstacle to obtain the corresponding obstacle avoidance prompt speech rate, and determining the obstacle avoidance prompt order based on the obstacle avoidance risk level of each obstacle; the second preset mapping relationship includes the correspondence between the obstacle avoidance risk level and the obstacle avoidance prompt speech rate; and providing obstacle avoidance voice prompts to the blind person based on the obstacle avoidance prompt text for each obstacle according to the obstacle avoidance prompt order, the obstacle avoidance prompt voice intensity, and the obstacle avoidance prompt speech rate for each obstacle.

[0089] In one embodiment, the obstacle avoidance module 340 determines the obstacle avoidance prompt voice intensity based on the current signal frequency and a preset noise main frequency band, including: when the current signal frequency is within the preset noise main frequency band, determining the obstacle avoidance prompt voice intensity as the sum of a preset baseline prompt voice intensity and a preset intensity increase; when the current signal frequency is outside the preset noise main frequency band, determining the obstacle avoidance prompt voice intensity as the preset baseline prompt voice intensity.

[0090] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is merely an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the functional modules described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0091] The obstacle avoidance navigation device for the blind provided in this embodiment can be applied to any of the obstacle avoidance navigation methods for the blind provided in the above embodiments, and has the corresponding functions and beneficial effects.

[0092] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 4 A block diagram is shown of an exemplary electronic device 11 suitable for implementing embodiments of the present application. Figure 4 The electronic device 11 shown is merely an example and should not impose any limitations on the functionality and scope of use of this embodiment.

[0093] like Figure 4 As shown, the electronic device 11 is represented in the form of a general-purpose computing electronic device. The components of the electronic device 11 may include, but are not limited to: one or more processors or processing units 16, system memory 28, and bus 18 connecting different system components (including system memory 28 and processing unit 16).

[0094] Bus 18 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. For example, these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0095] Electronic device 11 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by electronic device 11, including volatile and non-volatile media, removable and non-removable media.

[0096] System memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. Electronic device 11 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 34 may be used to read and write non-removable, non-volatile magnetic media (… Figure 4 Not shown; usually referred to as a "hard drive"). Although Figure 4As not shown, disk drives for reading and writing to removable non-volatile disks (e.g., "floppy disks") and optical disc drives for reading and writing to removable non-volatile optical discs (e.g., CD-ROMs, DVD-ROMs, or other optical media) may be provided. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. System memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of this application.

[0097] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 42 typically perform the functions and / or methods described in the embodiments of this application.

[0098] Electronic device 11 can also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), and with one or more devices that enable a user to interact with electronic device 11, and / or with any device that enables electronic device 11 to communicate with one or more other computing devices (e.g., network interface card and modem, etc.). This communication can be performed via input / output (I / O) interface 22. Furthermore, electronic device 11 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 20.

[0099] like Figure 4 As shown, network adapter 20 communicates with other modules of electronic device 11 via bus 18. It should be understood that, although... Figure 4 As not shown, other hardware and / or software modules may be used in conjunction with electronic device 11, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0100] The processing unit 16 executes various functional applications and page displays by running programs stored in the system memory 28, such as implementing a blind obstacle avoidance navigation method provided in any embodiment of this application.

[0101] This application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements, for example, a blind obstacle avoidance navigation method provided in any embodiment of this application.

[0102] The computer storage medium of this embodiment can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. For example, a computer-readable storage medium can be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer 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 device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0103] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0104] Program code contained on a computer-readable medium may be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0105] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0106] Those skilled in the art will understand that the modules or steps described above in this application can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, which can then be stored in a storage device for execution by a computing device. Alternatively, they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0107] Furthermore, the acquisition, storage, use, and processing of data in this application's technical solution all comply with relevant laws and regulations.

[0108] Note that the above are merely preferred embodiments and the technical principles employed in this application. Those skilled in the art will understand that this application is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of this application. Therefore, although this application has been described in detail through the above embodiments, this application is not limited to the above embodiments. Many other equivalent embodiments may be included without departing from the inventive concept of this application, and the scope of this application is determined by the scope of the appended claims.

Claims

1. A method for obstacle avoidance navigation for the blind, characterized in that, An intelligent control device applied to an obstacle avoidance navigation system for the blind, the method comprising: Acquire the current obstacle dataset, current environment image, and current voice data obtained at the current moment from a preset environmental area; the preset environmental area is an area centered on the blind person's location with a preset distance as its radius; Target detection is performed on the current environmental image to obtain the obstacle type of each obstacle. Based on the current warning distance and the obstacle type of each obstacle, a first preset mapping relationship is queried to obtain the obstacle weight factor of the corresponding obstacle. Based on the current obstacle data of each obstacle and the historical obstacle data of the corresponding obstacle, the speed sensitivity coefficient of the corresponding obstacle is determined. The current obstacle dataset includes the current obstacle data of each obstacle; the first preset mapping relationship includes the correspondence between warning distance, obstacle type and obstacle weight factor. The noise interference coefficient of the preset environment area at the current moment is determined based on the current signal strength in the current voice data, and the obstacle avoidance risk index of the corresponding obstacle is determined based on the noise interference coefficient, the current obstacle data of each obstacle, the obstacle weight factor and the speed sensitivity coefficient. Classified obstacle avoidance navigation for blind people is based on the obstacle avoidance risk index of each obstacle. The steps for determining the current warning distance are as follows: obtain the average reaction time of blind people over a historical period to obtain the historical average reaction time; determine the candidate distance compensation ratio based on the historical average reaction time, the preset baseline reaction time, and the preset compensation scaling factor, and determine the target distance compensation ratio based on the candidate distance compensation ratio and the preset maximum compensation coefficient; calculate the product of the target distance compensation ratio and the preset baseline warning distance to obtain the distance compensation amount, and calculate the sum of the preset baseline warning distance and the distance compensation amount to obtain the current warning distance.

2. The obstacle avoidance navigation method for the blind according to claim 1, characterized in that, Current obstacle data includes current relative speed and current relative distance. Based on the noise interference coefficient, current obstacle data for each obstacle, obstacle weighting factor, and speed sensitivity coefficient, the obstacle avoidance risk index for the corresponding obstacle is determined, including: For the current obstacle among all obstacles, calculate the product of the speed sensitivity coefficient of the current obstacle and the absolute value of the current relative speed of the current obstacle to obtain the target approach rate, and calculate the ratio between the target approach rate and the current relative distance of the current obstacle to obtain the dynamic risk index; The static risk index is obtained by multiplying the noise interference coefficient with the obstacle weight factor of the current obstacle. The dynamic risk index and the static risk index are then summed to obtain the obstacle avoidance risk index of the current obstacle.

3. The obstacle avoidance navigation method for the blind according to claim 1, characterized in that, Based on the obstacle avoidance risk index of each obstacle, a graded obstacle avoidance navigation system is implemented for blind people, including: Based on the obstacle avoidance risk index of each obstacle, the obstacle avoidance risk level of the corresponding obstacle is determined, and based on the obstacle avoidance risk level, the prompt semantic template of the corresponding obstacle is selected from at least two preset semantic templates; Based on the current obstacle data, obstacle type, and prompt semantic template of each obstacle, obstacle avoidance prompt text is generated for the corresponding obstacle, and obstacle avoidance navigation is provided for the blind based on the obstacle avoidance prompt text of each obstacle.

4. The obstacle avoidance navigation method for the blind according to claim 3, characterized in that, The current voice data also includes the current signal frequency, and obstacle avoidance navigation for the blind based on obstacle avoidance prompts for each obstacle, including: The intensity of the obstacle avoidance prompt voice is determined based on the current signal frequency and the preset noise dominant frequency band. Based on the obstacle avoidance risk level of each obstacle, the second preset mapping relationship is queried to obtain the obstacle avoidance prompt speed of the corresponding obstacle, and the order of obstacle avoidance prompts is determined based on the obstacle avoidance risk level of each obstacle; the second preset mapping relationship includes the correspondence between obstacle avoidance risk level and obstacle avoidance prompt speed. Based on the obstacle avoidance prompts, the intensity of the prompts, and the speaking speed of each obstacle, visually impaired individuals are given obstacle avoidance prompts in the order of the prompts, the intensity of the prompts, and the speaking speed of the prompts for each obstacle.

5. The obstacle avoidance navigation method for the blind according to claim 4, characterized in that, The obstacle avoidance prompt voice intensity is determined based on the current signal frequency and the preset dominant noise frequency band, including: When the current signal frequency is within the preset noise main frequency band, the obstacle avoidance prompt voice intensity is determined to be the sum of the preset baseline prompt voice intensity and the preset intensity increase. When the current signal frequency is outside the preset noise main frequency band, the obstacle avoidance prompt voice intensity is determined to be the preset benchmark prompt voice intensity.

6. A navigation device for blind people to avoid obstacles, characterized in that, An intelligent control device for use in a navigation system for blind people, the device comprising: The acquisition module is used to acquire the current obstacle dataset, current environment image, and current voice data obtained from the preset environmental area at the current moment; the preset environmental area is an area centered on the location of the blind person with a preset distance as the radius; The first determining module is used to perform target detection on the current environmental image, obtain the obstacle type of each obstacle, query the first preset mapping relationship based on the current warning distance and the obstacle type of each obstacle to obtain the obstacle weight factor of the corresponding obstacle, and determine the speed sensitivity coefficient of the corresponding obstacle based on the current obstacle data of each obstacle and the historical obstacle data of the corresponding obstacle. The current obstacle dataset includes the current obstacle data of each obstacle; the first preset mapping relationship includes the correspondence between the warning distance, obstacle type and obstacle weight factor. The second determining module is used to determine the noise interference coefficient of the preset environmental area at the current moment based on the current signal strength in the current voice data, and to determine the obstacle avoidance risk index of the corresponding obstacle based on the noise interference coefficient, the current obstacle data of each obstacle, the obstacle weight factor and the speed sensitivity coefficient. The obstacle avoidance module is used to provide graded obstacle avoidance navigation for blind people based on the obstacle avoidance risk index of each obstacle. The steps for determining the current warning distance are as follows: obtain the average reaction time of blind people over a historical period to obtain the historical average reaction time; determine the candidate distance compensation ratio based on the historical average reaction time, the preset baseline reaction time, and the preset compensation scaling factor, and determine the target distance compensation ratio based on the candidate distance compensation ratio and the preset maximum compensation coefficient; calculate the product of the target distance compensation ratio and the preset baseline warning distance to obtain the distance compensation amount, and calculate the sum of the preset baseline warning distance and the distance compensation amount to obtain the current warning distance.

7. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the obstacle avoidance navigation method for the blind as described in any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the obstacle avoidance navigation method for the blind as described in any one of claims 1 to 5.

Images