Automobile data acquisition method and system and medium

Abstract

The invention relates to an automobile data acquisition method and system and a medium. The method comprises the following steps: L1, acquiring data information of parameters of an automobile end and data information of parameters of a roadside end in real time; l2, based on the data information of the parameters of the vehicle end and the data information of the parameters of the roadside end, fusing the parameters of the vehicle end and the parameters of the roadside end by adopting an improved multi-modal feature fusion algorithm based on a covariance matrix to obtain the data information of the fused parameters, and representing the priority sequence of the fused parameters by adopting a multi-process scheduling algorithm based on a Spearman correlation coefficient, and constructing a priority matrix of the fused parameters to obtain data information of the priority matrix of the fused parameters. According to the method, the automobile data can be accurately and quickly acquired, the automobile can be conveniently and quickly adjusted, so that the safety of the automobile in the driving process is improved, the priority of the data is selectively acquired, and the overall running efficiency of the automobile is improved.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to an automobile data collection method, system and medium. Background Art

[0002] With the continuous development and progress of science and technology, driverless technology has gradually become an important research direction in the current scientific and technological field. The current driverless technology is based on a large amount of data training and simulation testing. Therefore, obtaining real driving data on the vehicle side is the key to the application and performance improvement of driverless buses.

[0003] However, the smart connected cars currently on the market can generally only transmit data back through the network of the on-board Internet of Things card provided by the operator. The data transmission is affected by the operator's network environment and the network uplink bandwidth. Its transmission rate cannot meet the transmission of large amounts of data. At the same time, considering that the data storage space on the vehicle side is limited and data related to perception, planning, positioning, etc. related to autonomous driving will continue to be generated during driving, a large amount of driving data cannot be collected in time and is cleared and overwritten, thus being completely lost.

[0004] Using mature wireless LAN technology, the driverless bus automatically connects to the wireless LAN of the platform or parking lot when it stops at a designated location such as a smart station or smart parking lot. Because the wireless LAN has a high network transmission rate and is stable and reliable, the amount of data that can be transmitted at each stop or parking lot can be calculated through the driverless bus operation schedule, so that the data that meets the conditions can be efficiently transmitted to the corresponding storage device on the roadside each time the LAN is connected. However, how to control and adjust the data collection during the data transmission process to improve the accuracy and convenience of data collection has become a problem that we need to solve urgently. Summary of the invention

[0005] In view of the above problems, the present invention provides a vehicle data collection method, system and medium, which can not only accurately and quickly collect vehicle data, facilitate rapid adjustment of the vehicle, thereby improving the safety of the vehicle driving process, but also selectively collect data priority to improve the overall operation efficiency of the vehicle.

[0006] In order to achieve the above-mentioned purpose and other related purposes, the technical solution provided by the present invention is as follows: a method for collecting automobile data, the method comprising:

[0007] L1. Collect data information of vehicle-side parameters and road-side parameters in real time;

[0008] L2. Based on the data information of the parameters of the vehicle side and the data information of the parameters of the road side, the parameters of the vehicle side and the parameters of the road side are fused using an improved multimodal feature fusion algorithm based on the covariance matrix to obtain the data information of the fused parameters;

[0009] L3. Based on the data information of the fused parameters, a multi-process scheduling algorithm based on the Spearman correlation coefficient is used to characterize the priority sequence of the fused parameters, and a priority matrix of the fused parameters is constructed to obtain data information of the priority matrix of the fused parameters;

[0010] L4. Based on the data information of the priority matrix of the fused parameters, an adaptively adjusted automobile data acquisition model is constructed to control and adjust the acquisition of the fused parameters, and output control data information of automobile data acquisition.

[0011] Furthermore, in step L2, the step of fusing the vehicle-side parameters and the roadside-side parameters using an improved multimodal feature fusion algorithm based on a covariance matrix includes:

[0012] L21. Based on the data information of the vehicle-side parameters, construct a feature extraction function Q of the vehicle-side parameters,

[0013]

[0014] Among them, x is the parameter data information of the vehicle side, α1, α2 and α3 are the feature extraction factors of the vehicle side parameters, and the features of the parameters of the vehicle side are extracted to obtain the feature data information of the parameters of the vehicle side;

[0015] L22. Based on the data information of the roadside end parameters, construct a feature extraction function W of the roadside end parameters,

[0016]

[0017] Wherein, y is the data information of the parameters of the roadside end, β1, β2 and β3 are the feature extraction factors of the parameters of the roadside end, and the features of the parameters of the roadside end are extracted to obtain the feature data information of the parameters of the roadside end;

[0018] L23. Based on the characteristic data information of the parameters of the roadside end and the characteristic data information of the parameters of the vehicle end, a multimodal feature fusion function R is established,

[0019]

[0020] Among them, r1 is the characteristic data information of the parameters on the road side, r2 is the characteristic data information of the parameters on the vehicle side, δ1, δ2 and δ3 are the covariance matrices of the road side parameters and the vehicle side parameters. The parameters on the vehicle side and the road side are fused to obtain the data information of the fused parameters.

[0021] Furthermore, the covariance matrices δ1, δ2 and δ3 of the roadside parameters and vehicle-side parameters are:

[0022]

[0023]

[0024] Among them, r1 is the characteristic data information of the parameters of the road side, and r2 is the characteristic data information of the parameters of the vehicle side.

[0025] Furthermore, the constraints of the feature extraction factors α1, α2 and α3 of the vehicle parameters are:

[0026]

[0027] The constraint function f of the feature extraction factors β1, β2 and β3 of the roadside end parameters is,

[0028]

[0029] The value range of the constraint function f is (0,1).

[0030] Furthermore, in step L3, the use of a multi-process scheduling algorithm based on the Spearman correlation coefficient to characterize the priority sequence of the fused parameters includes:

[0031] L31. Based on the data information of the fused parameters, a function P of the Spearman correlation coefficient of the parameters is established.

[0032]

[0033] Among them, z is the data information of the fused parameters, γ1, γ2 and γ3 are the weight factors of the parameters, and the Spearman correlation coefficient of the fused parameters is calculated to obtain the data information of the Spearman correlation coefficient of the fused parameters;

[0034] L32. Based on the data information of the Spearman correlation coefficient of the fused parameters, establish a priority scheduling function S of the parameters,

[0035]

[0036] Among them, h is the data information of the Spearman correlation coefficient of the fused parameters, η1, η2 and η3 are the priority scheduling factors of the fused parameters;

[0037] L33. Based on the priority scheduling function S of the parameters, the priority sequence of the fused parameters is characterized, and a priority matrix of the fused parameters is constructed to obtain data information of the priority matrix of the fused parameters.

[0038] Furthermore, the constraint function g1 of the weight factors γ1, γ2 and γ3 of the parameters is,

[0039] The constraint function g2 of the priority scheduling factor of the fused parameters is:

[0040]

[0041] Among them, the value range of constraint function g1 is (1,2), and the value range of constraint function g2 is (3,5).

[0042] Furthermore, in step L4, the construction of the adaptively adjusted automobile data collection model to control and adjust the collection of the fused parameters includes:

[0043] L41. Based on the data information of the priority matrix of the fused parameters, construct a priority data set of the fused parameters;

[0044] L42. The priority data set of the fused parameters is input into the adaptively adjusted vehicle data acquisition model for training and learning, and the control function G of the vehicle data acquisition is determined,

[0045] Among them, q is the priority data set of the fused parameters, λ1, λ2 and λ3 are the control factors of the adaptive adjustment of the vehicle data collection, and the trained adaptive adjustment vehicle data collection model is obtained;

[0046] L43. Based on the trained adaptively adjusted automobile data acquisition model, input the data information of the priority matrix of the fused parameters, control and adjust the acquisition of the fused parameters, and output the control data information of the automobile data acquisition.

[0047] Furthermore, the control factors λ1, λ2 and λ3 of the adaptive adjustment of the vehicle data collection are:

[0048]

[0049] Among them, q is the priority dataset of the fused parameters.

[0050] In order to achieve the above-mentioned object and other related objects, the present invention also provides an automobile data collection system, including a computer device, which is programmed or configured to execute the steps of any one of the above-mentioned automobile data collection methods.

[0051] In order to achieve the above-mentioned object and other related objects, the present invention further provides a computer-readable storage medium, on which is stored a computer program programmed or configured to execute any one of the above-mentioned vehicle data collection methods.

[0052] The present invention has the following positive effects:

[0053] 1. The present invention fuses the parameters of the vehicle side and the parameters of the road side by adopting an improved multimodal feature fusion algorithm based on the covariance matrix, and characterizes the priority sequence of the fused parameters in combination with a multi-process scheduling algorithm based on the Spearman correlation coefficient, and constructs a priority matrix of the fused parameters. It not only accurately analyzes the parameters of the vehicle and the road side, thereby ensuring the convenience of the automobile data collection process, but also the vehicle can process the data in sequence according to the priority of the transmitted data, thereby ensuring the safety of the vehicle operation process and preventing accidents.

[0054] 2. The present invention constructs an adaptively adjusted automobile data collection model to control and adjust the collection of fused parameters. It can not only accurately and quickly collect automobile data, facilitate rapid adjustment of the vehicle, thereby improving the safety of the vehicle's driving process, but also selectively collect data priorities to improve the overall efficiency of the vehicle's operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0056] Figure 2 It is a flow chart of an improved multi-modal feature fusion algorithm based on covariance matrix of the present invention;

[0057] Figure 3 It is a schematic diagram of the process of the multi-process scheduling algorithm based on the Spearman correlation coefficient of the present invention;

[0058] Figure 4 A schematic diagram of a process for constructing an adaptively adjusted automobile data collection model of the present invention;

[0059] Figure 5 It is a schematic diagram of the system framework of the present invention. DETAILED DESCRIPTION

[0060] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0061] Example 1: Figure 1 As shown, a method for collecting automobile data comprises:

[0062] L1. Collect data information of vehicle-side parameters and road-side parameters in real time;

[0063] L2. Based on the data information of the parameters of the vehicle side and the data information of the parameters of the road side, the parameters of the vehicle side and the parameters of the road side are fused using an improved multimodal feature fusion algorithm based on the covariance matrix to obtain the data information of the fused parameters;

[0064] L3. Based on the data information of the fused parameters, a multi-process scheduling algorithm based on the Spearman correlation coefficient is used to characterize the priority sequence of the fused parameters, and a priority matrix of the fused parameters is constructed to obtain data information of the priority matrix of the fused parameters;

[0065] L4. Based on the data information of the priority matrix of the fused parameters, an adaptively adjusted automobile data acquisition model is constructed to control and adjust the acquisition of the fused parameters, and output control data information of automobile data acquisition.

[0066] In this embodiment, if Figure 2 As shown, in step L2, the method of fusing the vehicle-side parameters and the roadside-side parameters using the improved multimodal feature fusion algorithm based on the covariance matrix includes:

[0067] L21. Based on the data information of the vehicle-side parameters, construct a feature extraction function Q of the vehicle-side parameters,

[0068]

[0069] Among them, x is the parameter data information of the vehicle side, α1, α2 and α3 are the feature extraction factors of the vehicle side parameters, and the features of the parameters of the vehicle side are extracted to obtain the feature data information of the parameters of the vehicle side;

[0070] L22. Based on the data information of the roadside end parameters, construct a feature extraction function W of the roadside end parameters,

[0071]

[0072] Wherein, y is the data information of the parameters of the roadside end, β1, β2 and β3 are the feature extraction factors of the parameters of the roadside end, and the features of the parameters of the roadside end are extracted to obtain the feature data information of the parameters of the roadside end;

[0073] L23. Based on the characteristic data information of the parameters of the roadside end and the characteristic data information of the parameters of the vehicle end, a multimodal feature fusion function R is established,

[0074]

[0075] Among them, r1 is the characteristic data information of the parameters on the road side, r2 is the characteristic data information of the parameters on the vehicle side, δ1, δ2 and δ3 are the covariance matrices of the road side parameters and the vehicle side parameters. The parameters on the vehicle side and the road side are fused to obtain the data information of the fused parameters.

[0076] In this embodiment, the covariance matrices δ1, δ2 and δ3 of the roadside parameters and the vehicle parameters are:

[0077]

[0078]

[0079] Among them, r1 is the characteristic data information of the parameters of the road side, and r2 is the characteristic data information of the parameters of the vehicle side.

[0080] In this embodiment, the constraints of the feature extraction factors α1, α2 and α3 of the vehicle parameters are:

[0081]

[0082] The constraint function f of the feature extraction factors β1, β2 and β3 of the roadside end parameters is,

[0083]

[0084] The value range of the constraint function f is (0,1).

[0085] In this embodiment, if Figure 3 As shown, in step L3, the use of a multi-process scheduling algorithm based on the Spearman correlation coefficient to characterize the priority sequence of the fused parameters includes:

[0086] L31. Based on the data information of the fused parameters, a function P of the Spearman correlation coefficient of the parameters is established.

[0087]

[0088] Among them, z is the data information of the fused parameters, γ1, γ2 and γ3 are the weight factors of the parameters, and the Spearman correlation coefficient of the fused parameters is calculated to obtain the data information of the Spearman correlation coefficient of the fused parameters;

[0089] L32. Based on the data information of the Spearman correlation coefficient of the fused parameters, establish a priority scheduling function S of the parameters,

[0090]

[0091] Among them, h is the data information of the Spearman correlation coefficient of the fused parameters, η1, η2 and η3 are the priority scheduling factors of the fused parameters;

[0092] L33. Based on the priority scheduling function S of the parameters, the priority sequence of the fused parameters is characterized, and a priority matrix of the fused parameters is constructed to obtain data information of the priority matrix of the fused parameters.

[0093] In this embodiment, the constraint function g1 of the weight factors γ1, γ2 and γ3 of the parameters is,

[0094]

[0095] The constraint function g2 of the priority scheduling factor of the fused parameters is:

[0096]

[0097] Among them, the value range of constraint function g1 is (1,2), and the value range of constraint function g2 is (3,5).

[0098] Embodiment 2: Based on the method, system and medium for collecting automobile data in Embodiment 1, the present invention is further illustrated and described below.

[0099] like Figure 1 As shown, a method for collecting automobile data comprises:

[0100] L1. Collect data information of vehicle-side parameters and road-side parameters in real time;

[0101] L2. Based on the data information of the parameters of the vehicle side and the data information of the parameters of the road side, the parameters of the vehicle side and the parameters of the road side are fused using an improved multimodal feature fusion algorithm based on the covariance matrix to obtain the data information of the fused parameters;

[0102] L3. Based on the data information of the fused parameters, a multi-process scheduling algorithm based on the Spearman correlation coefficient is used to characterize the priority sequence of the fused parameters, and a priority matrix of the fused parameters is constructed to obtain data information of the priority matrix of the fused parameters;

[0103] L4. Based on the data information of the priority matrix of the fused parameters, an adaptively adjusted automobile data acquisition model is constructed to control and adjust the acquisition of the fused parameters, and output control data information of automobile data acquisition.

[0104] In this embodiment, if Figure 4 As shown, in step L4, the construction of the adaptively adjusted automobile data acquisition model to control and adjust the acquisition of the fused parameters includes:

[0105] L41. Based on the data information of the priority matrix of the fused parameters, construct a priority data set of the fused parameters;

[0106] L42. The priority data set of the fused parameters is input into the adaptively adjusted vehicle data acquisition model for training and learning, and the control function G of the vehicle data acquisition is determined,

[0107]

[0108] Among them, q is the priority data set of the fused parameters, λ1, λ2 and λ3 are the control factors of the adaptive adjustment of the vehicle data collection, and the trained adaptive adjustment vehicle data collection model is obtained;

[0109] L43. Based on the trained adaptively adjusted automobile data acquisition model, input the data information of the priority matrix of the fused parameters, control and adjust the acquisition of the fused parameters, and output the control data information of the automobile data acquisition.

[0110] In this embodiment, the control factors λ1, λ2 and λ3 of the adaptive adjustment of the vehicle data collection are:

[0111]

[0112] Among them, q is the priority dataset of the fused parameters.

[0113] In this embodiment, the present invention provides an automobile data collection system, including a computer device, wherein the computer device is programmed or configured to execute the steps of any one of the above-mentioned automobile data collection methods.

[0114] like Figure 5 As shown in the figure, this system consists of three parts: vehicle-mounted terminal, roadside terminal, and vehicle-mounted built-in data return application system:

[0115] 1) In-vehicle networking terminal

[0116] The on-board networking terminal has the ability to obtain and cache vehicle autonomous driving data. The on-board networking terminal has the function of accessing the local area network WiFi. The on-board networking terminal receives abnormal event records issued by the platform, and marks the on-board logs near the time point corresponding to the abnormal event as high-priority upload files.

[0117] 2) Roadside terminal

[0118] The roadside terminal provides an externally encrypted wireless LAN signal through wireless LAN technology for access by the on-board networking terminals parked within the network coverage area. The roadside terminal has a certain amount of memory space and is connected to the Internet. The roadside terminal is installed in designated locations such as platforms and parking lots according to actual needs.

[0119] 3) In-vehicle built-in data backhaul application

[0120] The autonomous driving driving log is packaged and named, and the logs at abnormal times are marked with names. Priority uploading is set. The on-board networking terminal automatically identifies and accesses the wireless LAN belonging to different stop points, and measures the upload rate under the current network environment. The uploadable file size is calculated based on the known stop time.

[0121] Uploadable file size = stop time × upload rate. Receive priority upload instructions from the cloud platform, retrieve corresponding logs and mark the names, set priority upload, and successfully uploaded files will be cleared locally in the vehicle.

[0122] In this embodiment, the present invention provides a computer-readable storage medium, on which is stored a computer program programmed or configured to execute any one of the above-mentioned vehicle data collection methods.

[0123] Any reference to memory, storage, database or other media used in the embodiments provided in the present application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM) and memory bus dynamic RAM (RDRAM).

[0124] In summary, the present invention can not only accurately and quickly collect automobile data, facilitate rapid adjustment of the vehicle, thereby improving the safety of the vehicle's driving process, but also selectively collect data priorities to improve the overall efficiency of the vehicle's operation.

[0125] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for collecting automobile data, characterized in that: The method comprises: L1. Collect data information of vehicle-side parameters and road-side parameters in real time; L2. Based on the data information of the parameters of the vehicle side and the data information of the parameters of the road side, the parameters of the vehicle side and the parameters of the road side are fused using an improved multimodal feature fusion algorithm based on the covariance matrix to obtain the data information of the fused parameters; L3. Based on the data information of the fused parameters, a multi-process scheduling algorithm based on the Spearman correlation coefficient is used to characterize the priority sequence of the fused parameters, and a priority matrix of the fused parameters is constructed to obtain data information of the priority matrix of the fused parameters; L4. Based on the data information of the priority matrix of the fused parameters, an adaptively adjusted automobile data acquisition model is constructed to control and adjust the acquisition of the fused parameters, and output control data information of automobile data acquisition.

2. The method for collecting automobile data according to claim 1, characterized in that: In step L2, the method of fusing the vehicle-side parameters and the roadside-side parameters using an improved multimodal feature fusion algorithm based on a covariance matrix includes: L21. Based on the data information of the vehicle-side parameters, construct a feature extraction function Q of the vehicle-side parameters, Among them, x is the parameter data information of the vehicle side, α1, α2 and α3 are the feature extraction factors of the vehicle side parameters, and the features of the parameters of the vehicle side are extracted to obtain the feature data information of the parameters of the vehicle side; L22. Based on the data information of the roadside end parameters, construct a feature extraction function W of the roadside end parameters, Wherein, y is the data information of the parameters of the roadside end, β1, β2 and β3 are the feature extraction factors of the parameters of the roadside end, and the features of the parameters of the roadside end are extracted to obtain the feature data information of the parameters of the roadside end; L23. Based on the characteristic data information of the parameters of the roadside end and the characteristic data information of the parameters of the vehicle end, a multimodal feature fusion function R is established, Among them, r1 is the characteristic data information of the parameters on the road side, r2 is the characteristic data information of the parameters on the vehicle side, δ1, δ2 and δ3 are the covariance matrices of the road side parameters and the vehicle side parameters. The parameters on the vehicle side and the road side are fused to obtain the data information of the fused parameters.

3. The method for collecting automobile data according to claim 2, characterized in that: The covariance matrices δ1, δ2 and δ3 of the roadside parameters and vehicle parameters are: Among them, r1 is the characteristic data information of the parameters of the road side, and r2 is the characteristic data information of the parameters of the vehicle side.

4. The automobile data collection method according to claim 2, characterized in that: The constraints of the feature extraction factors α1, α2 and α3 of the vehicle parameters are: The constraint function f of the feature extraction factors β1, β2 and β3 of the roadside end parameters is, The value range of the constraint function f is (0,1).

5. The automobile data collection method according to claim 1, characterized in that: In step L3, the use of a multi-process scheduling algorithm based on the Spearman correlation coefficient to characterize the priority sequence of the fused parameters includes: L31. Based on the data information of the fused parameters, a function P of the Spearman correlation coefficient of the parameters is established. Among them, z is the data information of the fused parameters, γ1, γ2 and γ3 are the weight factors of the parameters, and the Spearman correlation coefficient of the fused parameters is calculated to obtain the data information of the Spearman correlation coefficient of the fused parameters; L32. Based on the data information of the Spearman correlation coefficient of the fused parameters, establish a priority scheduling function S of the parameters, Among them, h is the data information of the Spearman correlation coefficient of the fused parameters, η1, η2 and η3 are the priority scheduling factors of the fused parameters; L33. Based on the priority scheduling function S of the parameters, the priority sequence of the fused parameters is characterized, and a priority matrix of the fused parameters is constructed to obtain data information of the priority matrix of the fused parameters.

6. The automobile data collection method according to claim 5, characterized in that: The constraint function g1 of the weight factors γ1, γ2 and γ3 of the parameters is, The constraint function g2 of the priority scheduling factor of the fused parameters is: Among them, the value range of constraint function g1 is (1,2), and the value range of constraint function g2 is (3,5).

7. The automobile data collection method according to claim 1, characterized in that: In step L4, the construction of the adaptively adjusted automobile data collection model to control and adjust the collection of the fused parameters includes: L41. Based on the data information of the priority matrix of the fused parameters, construct a priority data set of the fused parameters; L42. The priority data set of the fused parameters is input into the adaptively adjusted vehicle data acquisition model for training and learning, and the control function G of the vehicle data acquisition is determined, Among them, q is the priority data set of the fused parameters, λ1, λ2 and λ3 are the control factors of the adaptive adjustment of the vehicle data collection, and the trained adaptive adjustment vehicle data collection model is obtained; L43. Based on the trained adaptively adjusted automobile data acquisition model, input the data information of the priority matrix of the fused parameters, control and adjust the acquisition of the fused parameters, and output the control data information of the automobile data acquisition.

8. The automobile data collection method according to claim 7, characterized in that: The control factors λ1, λ2 and λ3 of the adaptive adjustment of the vehicle data collection are, Among them, q is the priority dataset of the fused parameters.

9. An automobile data collection system, comprising a computer device, characterized in that: The computer device is programmed or configured to execute the steps of the vehicle data collection method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program programmed or configured to execute the vehicle data collection method according to any one of claims 1 to 8.

Images