Intelligent automobile driving control system based on visual analysis
Technical Field
The invention belongs to the field of driving control, and particularly relates to an intelligent automobile driving control system based on visual analysis.
Background
When driving a car, the driver can not be full of attention, always there is distraction, and the probability of accident when the driver distracts can greatly increase, the prior art usually identifies the distance between the front car or the surrounding car through a laser radar or a millimeter wave radar, if the distance is less than the safety distance, the accident is avoided through a mode of reminding the driver to decelerate or actively brake, but the cost of the mode is too high, the identification content is less, the distance between the cars can only be identified, and the vehicle is also very difficult to install in the later stage, and the peripheral situation of the vehicle is identified through 360 cameras in the mode, so as to assist the driver to drive safely, but the content that can be identified by the mode is increased, but the weather limitation is received, so the problem needs to be solved urgently.
Disclosure of Invention
The invention aims to provide an automobile intelligent driving control system based on visual analysis, which aims to solve the problems in the prior art.
In order to achieve the above object, the present invention provides an intelligent driving control system for an automobile based on visual analysis, comprising:
the system comprises an acquisition module, a processing module, a control module and a vehicle information acquisition module;
the acquisition module is used for acquiring images inside and outside the vehicle;
the processing module is used for processing the acquired images inside and outside the vehicle, acquiring instruction information and sending the instruction information to the control module;
the control module is used for controlling the running of the vehicle and carrying out early warning reminding on a driver based on the instruction information;
the vehicle information acquisition module is used for acquiring the running information of the vehicle.
Optionally, the acquisition module includes a first CCD camera and a second CCD camera, the first CCD camera and the second CCD camera are both installed on the interior rearview mirror, wherein the first CCD camera is used for shooting images outside the vehicle, and the second CCD camera is used for shooting images inside the vehicle.
Optionally, the processing module includes a preprocessing module and an identification module;
the preprocessing module is used for carrying out noise reduction and graying on all the acquired images to obtain an in-vehicle grayscale image and an out-vehicle grayscale image;
the identification model is used for classifying and identifying the content in the grayed image, and corresponding instruction information is generated based on the obtained identification result.
Optionally, when the preprocessing module preprocesses all the acquired images, denoising the acquired images by using a wavelet threshold denoising method, including: the method comprises the steps of firstly performing wavelet threshold processing on lifting wavelet transformation and then performing reconstruction, wherein a biorthogonal wavelet construction method is adopted for the lifting wavelet transformation, the biorthogonal wavelet construction method utilizes a prediction operator to determine high-frequency information and preliminarily determine low-frequency information, and then corrects the preliminarily determined low-frequency information through an update operator so as to determine the low-frequency information, and the method is specifically operated to preset a matlab tool bag in a preprocessing module for denoising.
Optionally, the working process of the identification module includes:
constructing an identification model based on a convolutional neural network;
inputting the recognition data in the recognition database into the recognition model for training;
inputting the interior gray scale image and the exterior gray scale image into the recognition model for recognition after training;
and generating corresponding instruction information based on the obtained identification result and sending the instruction information to the control module.
Optionally, the identification database includes a plurality of groups of images outside the vehicle, images inside the vehicle and correct identification results, which are preprocessed after actual shooting, and the identification results include: the traffic sign comprises the content of a traffic sign, a traffic indicator light, a violation camera, a speed bump, rainy days, snowy days, foggy days, a vehicle distance and whether a driver smokes or operates a mobile phone, wherein the vehicle distance is the vehicle distance between the vehicle and a vehicle in front of the vehicle distance.
Optionally, the process of generating corresponding instruction information based on the obtained identification result, and controlling the vehicle to run and performing early warning and reminding on the driver through the control module includes:
when the content of the traffic sign is a speed limit requirement, generating a speed upper limit instruction based on an actual speed limit numerical value, and setting a speed upper limit for a vehicle by the control module based on the speed upper limit instruction;
when the content of the traffic sign is speed limit releasing, a speed limit releasing instruction is generated, and the control module releases the upper limit of the speed of the vehicle based on the speed limit releasing instruction;
when the content of the traffic sign is a forbidden sign, generating a corresponding warning instruction, and generating corresponding character warning information by the control module based on the warning instruction to be displayed on a central control display to remind a driver;
when the identification result is that the deceleration strip or the traffic indicator light is a red light, a deceleration instruction is generated, and the control module controls the vehicle to decelerate based on the deceleration instruction;
generating a following distance lower limit instruction when the recognition result is the rainy day, the snowy day and the foggy day, and controlling the following distance by the control module based on the following distance lower limit instruction and preventing the vehicle from being braked emergently;
when the fact that a driver smokes is recognized in the running process of the vehicle, a smoking-free instruction is generated, the control module generates corresponding character warning information based on the smoking-free instruction and displays the character warning information on the central control display so as to remind the driver of not smoking during driving;
when the mobile phone is operated by a driver, a mobile phone operation request instruction is generated in the vehicle running process, and the control module generates corresponding character warning information based on the mobile phone operation request instruction and displays the character warning information on a central control display so as to remind the driver not to operate the mobile phone in the driving process;
the vehicle information acquisition module is used for acquiring the speed of the vehicle, and generating an over-close command of the distance of the vehicle based on the speed of the vehicle and the distance of the vehicle, and the control module is used for reminding a driver of over-close of the distance of the vehicle based on the over-close command of the distance of the vehicle;
and when the distance between the vehicles is reduced by more than 20m within 1s, generating an emergency braking instruction, and controlling the vehicles to perform emergency braking by the control module based on the emergency braking instruction.
Optionally, the generating of the too-close distance command based on the vehicle speed of the vehicle and the distance between the vehicle and the vehicle in front of the vehicle includes:
the speed of the vehicle is 80km/h-89km/h, and the distance is less than 90 m;
the speed of the vehicle is 90km/h-99km/h, and the distance between vehicles is less than 100 m;
the speed of the vehicle is 100km/h-109km/h, and the distance between vehicles is less than 110 m;
the speed of the vehicle is 110km/h-120km/h, and the distance between the vehicles is less than 120 m.
Optionally, the system further includes a meteorological system and a high-precision map module, where the meteorological system is configured to receive meteorological information of a meteorological website in real time, and assist the identification model in determining meteorological conditions;
the high-precision map module is used for connecting a map database on a line to acquire lane line data so as to assist the normal running of the vehicle in rainy days, snowy days and foggy days.
Optionally, the system further includes a mobile terminal, configured to remotely view vehicle driving information, where the vehicle driving information includes a single driving duration, a single driving oil consumption, a single driving track, weather during single driving, an average oil consumption, a driving average speed, and a peak speed.
The invention has the technical effects that:
according to the invention, through visual analysis, a plurality of conditions which are possibly required to be decelerated in front can be obtained in advance, and the deceleration is carried out in advance, so that the meteorological conditions can be identified, and thus, corresponding processing is carried out on different meteorological conditions, such as increasing the safe following distance and avoiding emergency braking, instructions are generated when the first camera identifies the contents of a traffic sign, whether a traffic signal lamp is a red light, whether a deceleration strip exists and the like, the vehicle is correspondingly controlled, and instructions are generated when the second camera identifies the conditions that a driver in the vehicle has driving smoke and plays a mobile phone, so that the driver in the vehicle is correspondingly reminded. Through discerning the condition in advance, increased the time of handling the condition, improved the security of driving, slowed down in advance to situations such as red light, deceleration strip and vehicle distance, improved fuel economy.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the application and, together with the description, serve to explain the application and are not intended to limit the application. In the drawings:
fig. 1 is a schematic structural diagram in an embodiment of the present invention.
Detailed Description
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.
It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer-executable instructions and that, although a logical order is illustrated in the flowcharts, in some cases, the steps illustrated or described may be performed in an order different than presented herein.
As shown in fig. 1, the present embodiment provides an intelligent driving control system for an automobile based on visual analysis, including:
the system comprises an acquisition module, a processing module, a control module and a vehicle information acquisition module;
the acquisition module is used for acquiring images inside and outside the vehicle;
the processing module is used for processing the collected images inside and outside the vehicle, acquiring instruction information and sending the instruction information to the control module;
the control module is used for controlling the running of the vehicle and carrying out early warning reminding on a driver based on the instruction information;
the vehicle information acquisition module is used for acquiring the running information of the vehicle.
Optionally, the acquisition module includes a first CCD camera and a second CCD camera, both the first CCD camera and the second CCD camera are mounted on the interior rearview mirror, wherein the first CCD camera is used for shooting images outside the vehicle, and the second CCD camera is used for shooting images inside the vehicle.
The processing module comprises a preprocessing module and an identification module;
the preprocessing module is used for carrying out noise reduction and graying on all the acquired images to obtain an in-vehicle grayscale image and an out-vehicle grayscale image;
the recognition model is used for classifying and recognizing the content in the grayscale image, and corresponding instruction information is generated based on the obtained recognition result.
When the preprocessing module preprocesses all the acquired images, denoising by a wavelet threshold denoising method comprises the following steps: the wavelet threshold processing is carried out on lifting wavelet transformation, then reconstruction is carried out, a biorthogonal wavelet construction method is adopted for lifting wavelet transformation, the biorthogonal wavelet construction method utilizes a prediction operator to determine high-frequency information and preliminarily determine low-frequency information, and then the preliminarily determined low-frequency information is corrected through an updating operator, so that the low-frequency information is determined, and the operation is specifically that a matlab toolkit is preset in a preprocessing module to carry out noise reduction.
The working process of the identification module comprises the following steps:
constructing an identification model based on a convolutional neural network;
inputting the recognition data in the recognition database into a recognition model for training;
inputting the interior gray scale image and the exterior gray scale image into the recognition model for recognition after training;
and generating corresponding instruction information based on the obtained identification result and sending the instruction information to the control module.
Optionally, the identification database includes a plurality of groups of images outside the vehicle, images inside the vehicle and correct identification results after actual shooting, and the identification results include: the content of the traffic sign, the traffic indicator light, the violation camera, the deceleration strip, the rainy day, the snowy day, the foggy day, the vehicle distance, whether the driver smokes or operates the mobile phone, and the vehicle distance is the vehicle distance between the vehicle and the front vehicle.
The process of generating corresponding instruction information based on the obtained identification result, controlling the running of the vehicle and carrying out early warning reminding on the driver through the control module comprises the following steps:
when the content of the traffic sign is a speed limit requirement, generating a speed upper limit instruction based on an actual speed limit numerical value, and setting a speed upper limit for the vehicle by the control module based on the speed upper limit instruction;
when the content of the traffic sign is speed limit releasing, a speed limit releasing instruction is generated, and the control module releases the upper limit of the speed of the vehicle based on the speed limit releasing instruction;
when the content of the traffic sign is a forbidden sign, generating a corresponding warning instruction, and generating corresponding text warning information by the control module based on the warning instruction to be displayed on the central control display to remind a driver;
when the identification result is that the deceleration strip or the traffic indicator light is a red light, a deceleration instruction is generated, and the control module controls the vehicle to decelerate based on the deceleration instruction;
generating a vehicle following distance lower limit instruction when the recognition result is rainy days, snowy days and foggy days, and controlling the vehicle following distance by the control module based on the vehicle following distance lower limit instruction and preventing the vehicle from being braked emergently;
in the case that the lower limit of the following distance defined by the following distance lower limit instruction is 50m, and when the visibility is lower than 50m due to weather, the control module does not intervene in control;
when the weather type is judged by identifying the road surface condition, classification can be realized by identifying the texture features of the ground, and the texture features of the ground are extracted based on a gray level co-occurrence matrix analysis method and a wavelet transform method.
In the driving process, a series of additional actions of taking cigarette, lighting cigarette and smoking by a driver all have certain potential safety hazards in driving: smoke-disturbed driving cabs can affect the driver's line of sight; the flames when lighting a cigarette can also cause the driver to be briefly dazzled at night. Meanwhile, in the case of tightly closing the windows of the vehicle, the driving smoke can cause the smoke in the cab to be distorted, the quality of air in the vehicle is extremely poor, the physical health of people in the vehicle is affected, and the people are even easy to nausea and aggravate dizziness. Therefore, the heavy smoke in the cab has certain danger for high-speed driving.
When the vehicle recognizes that the driver smokes, a smoking-free instruction is generated, and the control module generates corresponding character warning information based on the smoking-free instruction and displays the character warning information on the central control display to remind the driver not to smoke during driving;
when the mobile phone is operated by a driver, a mobile phone operation request instruction is generated in the vehicle running process, and the control module generates corresponding character warning information based on the mobile phone operation request instruction and displays the character warning information on the central control display so as to remind the driver not to operate the mobile phone in the driving process;
the method comprises the steps that the vehicle speed of a vehicle is obtained through a vehicle information acquisition module, an over-close command of the distance is generated based on the vehicle speed and the distance, and a control module reminds a driver of over-close of the distance based on the over-close command of the distance;
when the distance between vehicles is reduced by more than 20m within 1s, an emergency braking instruction is generated, and the control module controls the vehicles to perform emergency braking based on the emergency braking instruction.
The situation of generating the too-close distance command based on the vehicle speed and the distance between the vehicle and the front vehicle comprises the following steps:
the speed of the vehicle is 80km/h-89km/h, and the distance is less than 90 m;
the speed of the vehicle is 90km/h-99km/h, and the distance is less than 100 m;
the speed of the vehicle is 100km/h-109km/h, and the distance is less than 110 m;
the speed of the vehicle is 110km/h-120km/h, and the distance is less than 120 m.
The system also comprises a meteorological system and a high-precision map module, wherein the meteorological system is used for receiving meteorological information of a meteorological website in real time and assisting the identification model to judge meteorological conditions;
the high-precision map module is used for connecting a map database on a line to acquire lane line data so as to assist the normal running of the vehicle in rainy days, snowy days and foggy days.
The system also comprises a mobile terminal used for remotely checking the vehicle running information, wherein the vehicle running information comprises the single running time length, the single running oil consumption, the single running track, the weather during the single running, the average oil consumption, the running average speed and the peak speed.
The above description is only for the preferred embodiment of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present application should be covered within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.