CN117611569A - Vehicle fascia detection method, device, equipment and medium based on artificial intelligence - Google Patents
Vehicle fascia detection method, device, equipment and medium based on artificial intelligence Download PDFInfo
- Publication number
- CN117611569A CN117611569A CN202311685879.3A CN202311685879A CN117611569A CN 117611569 A CN117611569 A CN 117611569A CN 202311685879 A CN202311685879 A CN 202311685879A CN 117611569 A CN117611569 A CN 117611569A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- image
- fascia
- vehicle fascia
- damage
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 210000003195 fascia Anatomy 0.000 title claims abstract description 341
- 238000001514 detection method Methods 0.000 title claims abstract description 119
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 36
- 230000006378 damage Effects 0.000 claims abstract description 239
- 230000002787 reinforcement Effects 0.000 claims abstract description 61
- 238000012549 training Methods 0.000 claims abstract description 37
- 208000027418 Wounds and injury Diseases 0.000 claims description 72
- 208000014674 injury Diseases 0.000 claims description 72
- 210000002435 tendon Anatomy 0.000 claims description 26
- 230000015654 memory Effects 0.000 claims description 25
- 238000000034 method Methods 0.000 claims description 23
- 238000003708 edge detection Methods 0.000 claims description 16
- 230000006735 deficit Effects 0.000 claims description 12
- 238000004458 analytical method Methods 0.000 claims description 6
- 230000002708 enhancing effect Effects 0.000 claims description 4
- 230000000877 morphologic effect Effects 0.000 claims description 4
- 230000008569 process Effects 0.000 description 18
- 238000004422 calculation algorithm Methods 0.000 description 11
- 238000004590 computer program Methods 0.000 description 10
- 238000013528 artificial neural network Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 9
- 208000021945 Tendon injury Diseases 0.000 description 6
- 230000009466 transformation Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000010339 dilation Effects 0.000 description 4
- 238000001914 filtration Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000000844 transformation Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 239000003623 enhancer Substances 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000011158 quantitative evaluation Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
- G06N3/0455—Auto-encoder networks; Encoder-decoder networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
Abstract
The application provides a vehicle fascia detection method and device based on artificial intelligence, electronic equipment and storage medium, and the vehicle fascia detection method based on artificial intelligence comprises the following steps: collecting vehicle rib line images at different positions to obtain various types of vehicle rib line image sets; performing image enhancement on all images in the vehicle rib image set to obtain a rib image enhancement set; training a preset target detection network based on the reinforcement line image enhancement set to obtain a vehicle reinforcement line detection model; detecting the vehicle fascia image set based on the vehicle fascia detection model to obtain a vehicle fascia damage set; and determining damage to the damaged vehicle fascia based on the vehicle fascia damage set and a preset vehicle fascia standard image. According to the vehicle rib line detection method and device, the collected vehicle rib line images can be effectively enhanced, the enhanced images are utilized to train the vehicle rib line detection model, and therefore accuracy of vehicle rib line detection results is effectively improved.
Description
Technical Field
The application relates to the technical field of artificial intelligence, in particular to a vehicle fascia detection method, device, electronic equipment and storage medium based on artificial intelligence.
Background
In a common vehicle damage claim settlement activity, a surveyor needs to go to the site or a repair shop to examine the damage condition of the vehicle in the field, and then the problem of the claim settlement amount is negotiated and solved according to the claim settlement rules of the company and the self claim settlement experience. This process takes a lot of labor and time and is relatively inefficient for both the vehicle owner and the insurance company.
The detection of the vehicle fascia is a common and important part of the vehicle damage claim process, and the straight line or curve in the picture is usually detected by a Huffman algorithm in the prior art. However, the surface color of a general vehicle is single and smooth, the contrast is very low, and meanwhile, due to the fact that a natural scene is very complex, a smooth vehicle surface often has a reflection, a great error influence can be brought to a Huffman algorithm, and the accuracy of a vehicle fascia detection result is low.
Disclosure of Invention
In view of the foregoing, there is a need for an artificial intelligence-based vehicle fascia detection method, apparatus, electronic device, and storage medium that solve the technical problem of how to improve the accuracy of the vehicle fascia detection result.
The application provides a vehicle fascia detection method based on artificial intelligence, which comprises the following steps:
Collecting vehicle rib line images at different positions to obtain various types of vehicle rib line image sets;
performing image enhancement on all images in the vehicle rib image set to obtain a rib image enhancement set;
training a preset target detection network based on the reinforcement line image enhancement set to obtain a vehicle reinforcement line detection model;
detecting the vehicle fascia image set based on the vehicle fascia detection model to obtain a vehicle fascia damage set, wherein the vehicle fascia image set corresponds to the vehicle fascia damage set one by one;
and determining damage to the damaged vehicle fascia based on the vehicle fascia damage set and a preset vehicle fascia standard image.
In some embodiments, the acquiring vehicle rib image sets of different locations to obtain a plurality of categories of vehicle rib image sets includes:
acquiring vehicle fascia according to preset points to obtain vehicle fascia initial image sets of various categories, wherein the preset points correspond to the vehicle fascia initial image sets one by one;
and carrying out edge detection on all images in the vehicle fascia initial image set to obtain a plurality of types of vehicle fascia image sets.
In some embodiments, the image enhancing all images in the vehicle rib image set to obtain a rib image enhancement set includes:
Performing morphological operation on all images in the vehicle rib image set to obtain a rib image set;
and carrying out data enhancement on all images in the tendon image set to obtain a tendon image enhancement set.
In some embodiments, training the preset target detection network based on the reinforcement line image enhancement set to obtain a vehicle reinforcement line detection model includes:
setting labels for all images in the reinforcement line image enhancement set according to a preset mode to obtain a reinforcement line image label set, wherein the reinforcement line image enhancement set corresponds to the reinforcement line image label set one by one;
training a preset target detection network based on the fascia image enhancement set and the fascia image tag set to obtain a vehicle fascia detection model.
In some embodiments, the detecting the vehicle fascia image set based on the vehicle fascia detection model obtains a vehicle fascia injury set, the vehicle fascia image set and the vehicle fascia injury set being in one-to-one correspondence, including:
inputting all images in the vehicle fascia image set into the vehicle fascia detection model to obtain vehicle fascia damage pictures comprising a plurality of vehicle fascia damage types;
And taking the obtained vehicle fascia injury pictures comprising various vehicle fascia injury types as a vehicle fascia injury set, wherein the vehicle fascia image set corresponds to the vehicle fascia injury set one by one.
In some embodiments, the assigning damage to the damaged vehicle fascia based on the set of vehicle fascia damage and a preset vehicle fascia standard image comprises:
calculating the image similarity between each image in the vehicle rib line damage set and a preset vehicle rib line standard image;
calculating the damage assessment weight of each image in the vehicle rib line damage set;
carrying out weighted summation based on the loss assessment weight and the image similarity to obtain vehicle rib line damage degree, wherein the vehicle rib line damage degree corresponds to the vehicle rib line damage set one by one;
and determining damage to the damaged vehicle fascia based on the damage degree of the vehicle fascia.
In some embodiments, the calculating the impairment weights for each image in the set of vehicle fascia impairment comprises:
counting the total number of foreground pixels and the total number of background pixels of each image in the vehicle fascia injury set according to a connected domain analysis method;
taking the ratio of the total number of foreground pixels to the total number of background pixels of each image in the vehicle rib line damage set as the initial damage assessment weight corresponding to the image;
Normalizing the initial loss assessment weights of all the images in the vehicle rib line injury set to obtain the loss assessment weight of each image in the vehicle rib line injury set.
The embodiment of the application also provides a vehicle fascia detection device based on artificial intelligence, the device includes:
the acquisition unit is used for acquiring vehicle rib line images at different positions to obtain various types of vehicle rib line image sets;
the enhancement unit is used for carrying out image enhancement on all images in the vehicle rib image set to obtain a rib image enhancement set;
the training unit is used for training a preset target detection network based on the reinforcement line image enhancement set to obtain a vehicle reinforcement line detection model;
the detection unit is used for detecting the vehicle fascia image set based on the vehicle fascia detection model to obtain a vehicle fascia damage set, and the vehicle fascia image set corresponds to the vehicle fascia damage set one by one;
and the damage assessment unit is used for assessing damaged vehicle fascia based on the vehicle fascia damage set and a preset vehicle fascia standard image.
The embodiment of the application also provides electronic equipment, which comprises:
a memory storing at least one instruction;
And the processor executes the instructions stored in the memory to realize the vehicle fascia detection method based on artificial intelligence.
Embodiments of the present application also provide a computer readable storage medium having at least one instruction stored therein, the at least one instruction being executed by a processor in an electronic device to implement the artificial intelligence based vehicle fascia detection method.
According to the vehicle reinforcement line detection method and device, the obtained vehicle reinforcement line image is effectively enhanced, the preset target detection network is trained according to the enhanced image to obtain the vehicle reinforcement line detection model, and damaged vehicle reinforcement lines can be further subjected to damage assessment according to the vehicle reinforcement line damage set detected by the vehicle reinforcement line detection model, so that the accuracy of detecting the vehicle reinforcement line damage is improved.
Drawings
FIG. 1 is a flow chart of a preferred embodiment of an artificial intelligence based vehicle fascia detection method in accordance with the present application.
FIG. 2 is a functional block diagram of a preferred embodiment of an artificial intelligence based vehicle fascia detection apparatus in accordance with the present application.
Fig. 3 is a schematic structural diagram of an electronic device of a preferred embodiment of the artificial intelligence based vehicle fascia detection method according to the present application.
Detailed Description
In order that the objects, features and advantages of the present application may be more clearly understood, a more particular description of the invention will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings. It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application, the described embodiments are merely some, rather than all, of the embodiments of the present application.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more of the described features. In the description of the present application, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein in the description of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.
The embodiment of the application provides a vehicle fascia detection method based on artificial intelligence, which can be applied to one or more electronic devices, wherein the electronic devices are devices capable of automatically performing numerical calculation and/or information processing according to preset or stored instructions, and the hardware comprises, but is not limited to, a microprocessor, an application specific integrated circuit (Application Specific Integrated Circuit, an ASIC), a programmable gate array (Field-Programmable Gate Array, FPGA), a digital processor (Digital Signal Processor, DSP), an embedded device and the like.
The electronic device may be any electronic product that can interact with a customer in a human-machine manner, such as a personal computer, tablet, smart phone, personal digital assistant (Personal Digital Assistant, PDA), gaming machine, interactive web television (Internet Protocol Television, IPTV), smart wearable device, etc.
The electronic device may also include a network device and/or a client device. Wherein the network device includes, but is not limited to, a single network server, a server group composed of a plurality of network servers, or a Cloud based Cloud Computing (Cloud Computing) composed of a large number of hosts or network servers.
The network in which the electronic device is located includes, but is not limited to, the internet, a wide area network, a metropolitan area network, a local area network, a virtual private network (Virtual Private Network, VPN), and the like.
As shown in FIG. 1, a flow chart of a preferred embodiment of the artificial intelligence based vehicle fascia detection method of the present application is shown. The order of the steps in the flowchart may be changed and some steps may be omitted according to various needs.
S10, acquiring vehicle rib line images at different positions to obtain various types of vehicle rib line image sets.
In an alternative embodiment, the acquiring vehicle rib image sets of different locations to obtain a plurality of categories of vehicle rib image sets includes:
s101, acquiring vehicle fascia according to preset points to obtain various types of vehicle fascia initial image sets, wherein the preset points correspond to the vehicle fascia initial image sets one by one;
s102, carrying out edge detection on all images in the vehicle fascia initial image set to obtain a plurality of types of vehicle fascia image sets.
In this alternative embodiment, the RGB camera may be manually held to perform the same number of image acquisitions on the tendons at each location of the vehicle at the preset points, for example, 20 vehicle tendons at the corresponding locations are acquired at each preset point, and 20 vehicle tendon images corresponding to each preset point are used as the initial image set of the vehicle tendons corresponding to the point.
In this optional embodiment, the positions of the vehicle fascia corresponding to the preset points may be fascia corresponding to the hood, the fender, the bumper and the front, rear, left and right doors of the vehicle, so that in this solution, a conventional four-door vehicle type is taken as an example, at least seven preset points are included, and at least seven types of vehicle fascia initial image sets are correspondingly obtained.
In this alternative embodiment, all images in the vehicle fascia initial image set are first converted into gray-scale images, and then edge detection can be performed on the gray-scale images corresponding to all images in the vehicle fascia initial image set according to a Canny edge detection algorithm, so as to obtain edge lines on each image in the vehicle fascia initial image set.
In this alternative embodiment, canny edge detection is a standard image edge detection algorithm, and the main process of edge detection on all images in the vehicle fascia initial image set is as follows:
a) Multiplying each pixel point and the neighborhood pixel points in the gray level image by using a Gaussian matrix, and taking the calculated average value with weight as an updated gray level value of the pixel point;
b) Multiplying a pixel point in the gray image, which obtains an updated gray value, by a sobel operator to obtain gradient values of the pixel point in the horizontal and vertical directions;
c) Filtering non-maximum values, namely filtering points which are not edges in the gray level image, enabling the width of the edges to be as large as possible as one pixel point, if one pixel point belongs to the edges, enabling the gradient value of the pixel point in the horizontal or vertical direction to be maximum, otherwise, enabling the pixel not to be located at the edges, and setting the pixel value of the pixel to be 0;
d) Setting two thresholds of a maximum threshold and a minimum threshold, attributing pixels corresponding to all updated gray values larger than the maximum threshold in the gray image to edges, attributing pixels corresponding to all updated gray values smaller than the minimum threshold in the gray image to non-edges, attributing pixels corresponding to all updated gray values in the minimum threshold and the maximum threshold to edges if pixels are adjacent to edge pixels, and attributing pixels not adjacent to edge pixels to non-edges.
In this optional embodiment, all edge images obtained after edge detection of all images in each of the vehicle fascia initial image sets are used as a vehicle fascia image set corresponding to the vehicle fascia initial image set, that is, the vehicle fascia initial image set and the vehicle fascia image set are in one-to-one correspondence.
Therefore, the edge image of each acquired vehicle fascia image can be initially acquired, and data support is provided for detecting the vehicle fascia in the follow-up process.
S11, performing image enhancement on all images in the vehicle rib image set to obtain a rib image enhancement set.
In an alternative embodiment, said image enhancing all images in said vehicle rib image set to obtain a rib image enhancement set comprises:
s111, performing morphological operation on all images in the vehicle rib image set to obtain a rib image set;
and S112, carrying out data enhancement on all images in the tendon form image set to obtain a tendon form image enhancement set.
In this alternative embodiment, since the proportion of the pixels occupied by the lines in the vehicle line image set on the image is smaller, the edge lines of all the images in the vehicle line image set may be expanded by performing a dilation operation in morphology on all the images in the vehicle line image set first, where the convolution kernel of the dilation operation may be 5*5.
In the optional embodiment, in order to solve the problem of complex vehicle fascia damage scene in real life and improve the applicability and accuracy of a vehicle fascia detection model obtained by subsequent training, in the scheme, data enhancement is performed on all images in a fascia image set obtained after expansion operation.
In this alternative embodiment, all images in the set of rib-form images may be data enhanced by using a data enhancer such as Crop. The specific process may be to perform geometric transformations such as inversion, translation, rotation, scaling, cropping, etc., and color transformations on all images in each of the rib-form image sets; the color transformation, such as changing the gray level, brightness, or adding some noise to all images in the rib line image set. In the scheme, all images in the tendon form image set after data enhancement are used as tendon form image enhancement sets.
Therefore, the subsequent process is convenient to train by utilizing the reinforcement line image enhancement set obtained after data enhancement to obtain a more stable and effective vehicle reinforcement line detection model, and the accuracy of the detection result of the vehicle reinforcement line is improved.
And S12, training a preset target detection network based on the reinforcement line image enhancement set to obtain a vehicle reinforcement line detection model.
In an optional embodiment, the training the preset target detection network based on the reinforcement line image enhancement set to obtain the vehicle reinforcement line detection model includes:
s121, setting labels for all images in the reinforcement line image enhancement set according to a preset mode to obtain a reinforcement line image label set, wherein the reinforcement line image enhancement set corresponds to the reinforcement line image label set one by one;
S122, training a preset target detection network based on the reinforcement line image enhancement set and the reinforcement line image tag set to obtain a vehicle reinforcement line detection model.
In this optional embodiment, since all the images in the reinforcement set of the rib line image include not only the vehicle rib line but also the edge line of other parts or regions of the vehicle, all the edges included in all the images in the reinforcement set of the rib line image may be marked manually, and for all the edge portions belonging to the rib line of the vehicle in the images, different encoding labels may be set according to different damage types, for example, the encoding labels may be set according to the order of natural numbers, and all the regions not belonging to the rib line of the vehicle may be marked as 0, so as to obtain a rib line label image, and all the images in the reinforcement set of the rib line image after the label setting may be used as a rib line image label set.
In this optional embodiment, the preset target detection network may be a deephbv 3 neural network, and the network structure is an Encoder-Decoder structure, where the feature extraction is performed on the image in the reinforcement set of the input tendon image mainly by the Encoder, and the extracted features are fused by the Decoder to obtain the output image.
In this alternative embodiment, in order to enable the obtained vehicle fascia detection model to detect various types of vehicle fascia damage, the deep labv3 neural network needs to be trained first to obtain the vehicle fascia detection model, where the training process is the same as that of the existing target detection networks such as YOLO and FCOS. The main process is as follows: training loss between the output image and the corresponding rib line label image in the rib line image label set can be calculated through a cross entropy loss function, the deep LabV3 neural network is adjusted according to the training loss, so that the training loss is gradually reduced in a plurality of training processes, when the training loss is reduced to 0, the end of the training process is indicated, and the deep LabV3 neural network trained at the moment is used as a vehicle rib line detection model in the scheme.
Therefore, a more accurate vehicle fascia detection model can be obtained through a neural network training mode, and the accuracy of detecting the vehicle fascia damage in the subsequent process is improved.
And S13, detecting the vehicle fascia image set based on the vehicle fascia detection model to obtain a vehicle fascia damage set, wherein the vehicle fascia image set corresponds to the vehicle fascia damage set one by one.
In an optional embodiment, the detecting the vehicle fascia image set based on the vehicle fascia detection model to obtain a vehicle fascia damage set, where the vehicle fascia image set and the vehicle fascia damage set are in one-to-one correspondence, includes:
s131, inputting all images in the vehicle fascia image set into the vehicle fascia detection model to obtain vehicle fascia damage pictures comprising a plurality of vehicle fascia damage types;
and S132, taking the obtained vehicle fascia injury pictures comprising various vehicle fascia injury types as a vehicle fascia injury set, wherein the vehicle fascia image set corresponds to the vehicle fascia injury set one by one.
In this optional embodiment, all the images in the vehicle fascia image set are input into the trained vehicle fascia detection model, and the vehicle fascia detection model can sequentially detect the input images, so as to identify common damage such as fascia breakage, protrusion, recess, deformation and the like, which are common to the vehicle fascia.
In this optional embodiment, each type of vehicle rib line image set may correspond to vehicle rib line damage pictures of multiple types of vehicle rib line damage, and in this scheme, all types of vehicle rib line damage pictures corresponding to each type of vehicle rib line image set are used as vehicle rib line damage sets corresponding to the vehicle rib line image set.
Therefore, the rapid detection of all images in the vehicle rib line image set can be realized according to the vehicle rib line detection model, and the detection efficiency of the vehicle rib line damage is improved.
S14, damage is determined to the damaged vehicle fascia based on the vehicle fascia damage set and a preset vehicle fascia standard image.
In an alternative embodiment, said assigning damage to the damaged vehicle fascia based on the set of vehicle fascia damage and a preset vehicle fascia standard image includes:
s141, calculating the image similarity between each image in the vehicle fascia injury set and a preset vehicle fascia standard image;
s142, calculating the damage weight of each image in the vehicle fascia damage set;
s143, carrying out weighted summation based on the loss assessment weight and the image similarity to obtain vehicle fascia injury degree, wherein the vehicle fascia injury degree corresponds to the vehicle fascia injury set one by one;
s144, damage is determined to the damaged vehicle fascia based on the vehicle fascia damage degree.
In this optional embodiment, for the vehicle fascia damage set corresponding to each preset point location, a vehicle fascia standard image corresponding to the point location may be obtained in advance under the condition that the vehicle fascia is not damaged. And then respectively calculating the image similarity between each image in the vehicle fascia injury set and a preset vehicle fascia standard image according to a normalized cross-correlation matching algorithm.
In this optional embodiment, the normalized cross-correlation matching algorithm uses the vehicle fascia standard image as a template, and obtains the image similarity between each image in the vehicle fascia damage set and a preset vehicle fascia standard image by traversing each pixel of each image in the vehicle fascia damage set and comparing whether each pixel is similar to the template, wherein the value range is [0,1], and the closer 1 indicates the higher the similarity.
In this alternative embodiment, since the vehicle fascia injury set corresponding to each preset point location may include a plurality of vehicle fascia injury pictures, the vehicle fascia injury degree may be obtained by calculating an impairment weight of each image in the vehicle fascia injury set and performing weighted summation based on the impairment weight and the image similarity.
In this alternative embodiment, the total number of foreground pixels and the total number of background pixels of each image in the vehicle tendon injury set may be counted first according to a connected domain analysis method, where the connected domain analysis method is used to find and mark adjacent pixels with the same pixel value in the image. The foreground pixel total number refers to the pixel total number corresponding to the vehicle fascia region in each image in the vehicle fascia injury set, and the background pixel total number is the total number of all pixels excluding the vehicle fascia region in each image in the vehicle fascia injury set.
In this optional embodiment, the ratio of the total number of foreground pixels to the total number of background pixels of each image in the vehicle fascia injury set is used as the initial injury assessment weight corresponding to the image; and normalizing the initial loss assessment weights of all the images in the vehicle rib line injury set to obtain the loss assessment weight of each image in the vehicle rib line injury set.
For example, there are A, B, C, D images in the vehicle rib line injury set, and each image corresponds to an initial loss assessment weight of 0.2, 0.12, 0.16 and 0.32, and then the loss assessment weights obtained by normalizing the initial loss assessment weights of the four images are respectively 0.25, 0.15, 0.2 and 0.4.
In this alternative embodiment, the weighted summation may be performed based on the impairment weight of each image in the vehicle fascia injury set and the corresponding image similarity, so as to obtain the overall similarity between all the images in the vehicle fascia injury set and the vehicle fascia standard image, and the overall similarity is used as the vehicle fascia injury degree of the vehicle fascia at the corresponding point of the vehicle fascia injury set.
For example, there are A, B, C, D images in the vehicle fascia injury set, and the similarity between the images and the vehicle fascia standard image is 0.4, 0.8, 0.3, and 0.5, and the corresponding impairment weights are 0.25, 0.15, 0.2, and 0.4, respectively, so that the final calculated vehicle fascia injury is 0.4x0.25+0.8x0.15+0.3x0.2+0.5x0.4=0.48.
In this alternative embodiment, the historical data of all the vehicle tendon injury degrees corresponding to the vehicle tendons at each point location and the damage assessment line corresponding to each vehicle tendon injury degree may be stored in the database. When the latest obtained vehicle rib line damage degree is subjected to damage assessment, the average value of all the vehicle rib line damage degrees corresponding to the vehicle rib line at the current point position and the average value of the damage assessment amount corresponding to each vehicle rib line damage degree can be used as a standard damage assessment data pair, and the damage assessment amount corresponding to the current vehicle rib line damage degree can be calculated by referring to the standard damage assessment data pair.
For example, there are 4 damage conditions in the vehicle fascia at the vehicle hood position in the history, the damage degree of each vehicle fascia is respectively 0.5, 0.6, 0.2 and 0.3, the corresponding damage limit is 5000, 8000, 3000 and 4000, the average value of the damage degree of the vehicle fascia is calculated to obtain 0.4, the average value of the damage limit is calculated to obtain 5000, 8000, 3000 and 4000 is calculated to obtain 5000, and the corresponding standard damage data pair is (0.4 and 5000). If the current calculated vehicle fascia injury degree is 0.48, the current vehicle fascia injury degree corresponds to an estimated injury limit of 0.48/0.4x5000=6000.
In this optional embodiment, after the current vehicle rib line damage degree and the corresponding damage assessment limit are obtained through each calculation, the standard damage assessment data pair can be updated, so that when the vehicle rib line damage is assessed, a vehicle claim settlement person can refer to the historical data corresponding to all damage assessment limits of the vehicle rib line at the current point location, and more objective and fair judgment can be made on the current vehicle rib line damage to obtain more accurate damage assessment limit.
Therefore, the vehicle reinforcement line damage degree of each point position can be calculated to carry out overall quantitative evaluation on the vehicle reinforcement line, and the damage assessment efficiency and accuracy of insurance claim personnel on the vehicle reinforcement line damage are improved.
Referring to fig. 2, fig. 2 is a functional block diagram of a preferred embodiment of the artificial intelligence based vehicle fascia detection apparatus of the present application. The artificial intelligence based vehicle fascia detection apparatus 11 includes an acquisition unit 110, an enhancement unit 111, a training unit 112, a detection unit 113, and a damage assessment unit 114. The module/unit referred to herein is a series of computer readable instructions capable of being executed by the processor 13 and of performing a fixed function, stored in the memory 12. In the present embodiment, the functions of the respective modules/units will be described in detail in the following embodiments.
In an alternative embodiment, the acquisition unit 110 is configured to acquire vehicle rib image sets of various types by acquiring vehicle rib image sets of different locations.
In an alternative embodiment, the acquiring vehicle rib image sets of different locations to obtain a plurality of categories of vehicle rib image sets includes:
acquiring vehicle fascia according to preset points to obtain vehicle fascia initial image sets of various categories, wherein the preset points correspond to the vehicle fascia initial image sets one by one;
and carrying out edge detection on all images in the vehicle fascia initial image set to obtain a plurality of types of vehicle fascia image sets.
In this alternative embodiment, the RGB camera may be manually held to perform the same number of image acquisitions on the tendons at each location of the vehicle at the preset points, for example, 20 vehicle tendons at the corresponding locations are acquired at each preset point, and 20 vehicle tendon images corresponding to each preset point are used as the initial image set of the vehicle tendons corresponding to the point.
In this optional embodiment, the positions of the vehicle fascia corresponding to the preset points may be fascia corresponding to the hood, the fender, the bumper and the front, rear, left and right doors of the vehicle, so that in this solution, a conventional four-door vehicle type is taken as an example, at least seven preset points are included, and at least seven types of vehicle fascia initial image sets are correspondingly obtained.
In this alternative embodiment, all images in the vehicle fascia initial image set are first converted into gray-scale images, and then edge detection can be performed on the gray-scale images corresponding to all images in the vehicle fascia initial image set according to a Canny edge detection algorithm, so as to obtain edge lines on each image in the vehicle fascia initial image set.
In this alternative embodiment, canny edge detection is a standard image edge detection algorithm, and the main process of edge detection on all images in the vehicle fascia initial image set is as follows:
a) Multiplying each pixel point and the neighborhood pixel points in the gray level image by using a Gaussian matrix, and taking the calculated average value with weight as an updated gray level value of the pixel point;
b) Multiplying a pixel point in the gray image, which obtains an updated gray value, by a sobel operator to obtain gradient values of the pixel point in the horizontal and vertical directions;
c) Filtering non-maximum values, namely filtering points which are not edges in the gray level image, enabling the width of the edges to be as large as possible as one pixel point, if one pixel point belongs to the edges, enabling the gradient value of the pixel point in the horizontal or vertical direction to be maximum, otherwise, enabling the pixel not to be located at the edges, and setting the pixel value of the pixel to be 0;
d) Setting two thresholds of a maximum threshold and a minimum threshold, attributing pixels corresponding to all updated gray values larger than the maximum threshold in the gray image to edges, attributing pixels corresponding to all updated gray values smaller than the minimum threshold in the gray image to non-edges, attributing pixels corresponding to all updated gray values in the minimum threshold and the maximum threshold to edges if pixels are adjacent to edge pixels, and attributing pixels not adjacent to edge pixels to non-edges.
In this optional embodiment, all edge images obtained after edge detection of all images in each of the vehicle fascia initial image sets are used as a vehicle fascia image set corresponding to the vehicle fascia initial image set, that is, the vehicle fascia initial image set and the vehicle fascia image set are in one-to-one correspondence.
In an alternative embodiment, the enhancement unit 111 is configured to perform image enhancement on all images in the vehicle rib image set to obtain a rib image enhancement set.
In an alternative embodiment, said image enhancing all images in said vehicle rib image set to obtain a rib image enhancement set comprises:
Performing morphological operation on all images in the vehicle rib image set to obtain a rib image set;
and carrying out data enhancement on all images in the tendon image set to obtain a tendon image enhancement set.
In this alternative embodiment, since the proportion of the pixels occupied by the lines in the vehicle line image set on the image is smaller, the edge lines of all the images in the vehicle line image set may be expanded by performing a dilation operation in morphology on all the images in the vehicle line image set first, where the convolution kernel of the dilation operation may be 5*5.
In the optional embodiment, in order to solve the problem of complex vehicle fascia damage scene in real life and improve the applicability and accuracy of a vehicle fascia detection model obtained by subsequent training, in the scheme, data enhancement is performed on all images in a fascia image set obtained after expansion operation.
In this alternative embodiment, all images in the set of rib-form images may be data enhanced by using a data enhancer such as Crop. The specific process may be to perform geometric transformations such as inversion, translation, rotation, scaling, cropping, etc., and color transformations on all images in each of the rib-form image sets; the color transformation, such as changing the gray level, brightness, or adding some noise to all images in the rib line image set. In the scheme, all images in the tendon form image set after data enhancement are used as tendon form image enhancement sets.
In an alternative embodiment, the training unit 112 is configured to train the preset target detection network based on the reinforcement set of tendon image to obtain a vehicle tendon detection model.
In an optional embodiment, the training the preset target detection network based on the reinforcement line image enhancement set to obtain the vehicle reinforcement line detection model includes:
setting labels for all images in the reinforcement line image enhancement set according to a preset mode to obtain a reinforcement line image label set, wherein the reinforcement line image enhancement set corresponds to the reinforcement line image label set one by one;
training a preset target detection network based on the fascia image enhancement set and the fascia image tag set to obtain a vehicle fascia detection model.
In this optional embodiment, since all the images in the reinforcement set of the rib line image include not only the vehicle rib line but also the edge line of other parts or regions of the vehicle, all the edges included in all the images in the reinforcement set of the rib line image may be marked manually, and for all the edge portions belonging to the rib line of the vehicle in the images, different encoding labels may be set according to different damage types, for example, the encoding labels may be set according to the order of natural numbers, and all the regions not belonging to the rib line of the vehicle may be marked as 0, so as to obtain a rib line label image, and all the images in the reinforcement set of the rib line image after the label setting may be used as a rib line image label set.
In this optional embodiment, the preset target detection network may be a deephbv 3 neural network, and the network structure is an Encoder-Decoder structure, where the feature extraction is performed on the image in the reinforcement set of the input tendon image mainly by the Encoder, and the extracted features are fused by the Decoder to obtain the output image.
In this alternative embodiment, in order to enable the obtained vehicle fascia detection model to detect various types of vehicle fascia damage, the deep labv3 neural network needs to be trained first to obtain the vehicle fascia detection model, where the training process is the same as that of the existing target detection networks such as YOLO and FCOS. The main process is as follows: training loss between the output image and the corresponding rib line label image in the rib line image label set can be calculated through a cross entropy loss function, the deep LabV3 neural network is adjusted according to the training loss, so that the training loss is gradually reduced in a plurality of training processes, when the training loss is reduced to 0, the end of the training process is indicated, and the deep LabV3 neural network trained at the moment is used as a vehicle rib line detection model in the scheme.
In an alternative embodiment, the detecting unit 113 is configured to detect the vehicle fascia image set based on the vehicle fascia detection model to obtain a vehicle fascia damage set, where the vehicle fascia image set and the vehicle fascia damage set are in one-to-one correspondence.
In an optional embodiment, the detecting the vehicle fascia image set based on the vehicle fascia detection model to obtain a vehicle fascia damage set, where the vehicle fascia image set and the vehicle fascia damage set are in one-to-one correspondence, includes:
inputting all images in the vehicle fascia image set into the vehicle fascia detection model to obtain vehicle fascia damage pictures comprising a plurality of vehicle fascia damage types;
and taking the obtained vehicle fascia injury pictures comprising various vehicle fascia injury types as a vehicle fascia injury set, wherein the vehicle fascia image set corresponds to the vehicle fascia injury set one by one.
In this optional embodiment, all the images in the vehicle fascia image set are input into the trained vehicle fascia detection model, and the vehicle fascia detection model can sequentially detect the input images, so as to identify common damage such as fascia breakage, protrusion, recess, deformation and the like, which are common to the vehicle fascia.
In this optional embodiment, each type of vehicle rib line image set may correspond to vehicle rib line damage pictures of multiple types of vehicle rib line damage, and in this scheme, all types of vehicle rib line damage pictures corresponding to each type of vehicle rib line image set are used as vehicle rib line damage sets corresponding to the vehicle rib line image set.
In an alternative embodiment, the damage determination unit 114 is configured to determine damage to the damaged vehicle fascia based on the set of vehicle fascia damage and a predetermined vehicle fascia standard image.
In an alternative embodiment, said assigning damage to the damaged vehicle fascia based on the set of vehicle fascia damage and a preset vehicle fascia standard image includes:
calculating the image similarity between each image in the vehicle rib line damage set and a preset vehicle rib line standard image;
calculating the damage assessment weight of each image in the vehicle rib line damage set;
carrying out weighted summation based on the loss assessment weight and the image similarity to obtain vehicle rib line damage degree, wherein the vehicle rib line damage degree corresponds to the vehicle rib line damage set one by one;
and determining damage to the damaged vehicle fascia based on the damage degree of the vehicle fascia.
In this optional embodiment, for the vehicle fascia damage set corresponding to each preset point location, a vehicle fascia standard image corresponding to the point location may be obtained in advance under the condition that the vehicle fascia is not damaged. And then respectively calculating the image similarity between each image in the vehicle fascia injury set and a preset vehicle fascia standard image according to a normalized cross-correlation matching algorithm.
In this optional embodiment, the normalized cross-correlation matching algorithm uses the vehicle fascia standard image as a template, and obtains the image similarity between each image in the vehicle fascia damage set and a preset vehicle fascia standard image by traversing each pixel of each image in the vehicle fascia damage set and comparing whether each pixel is similar to the template, wherein the value range is [0,1], and the closer 1 indicates the higher the similarity.
In this alternative embodiment, since the vehicle fascia injury set corresponding to each preset point location may include a plurality of vehicle fascia injury pictures, the vehicle fascia injury degree may be obtained by calculating an impairment weight of each image in the vehicle fascia injury set and performing weighted summation based on the impairment weight and the image similarity.
In this alternative embodiment, the total number of foreground pixels and the total number of background pixels of each image in the vehicle tendon injury set may be counted first according to a connected domain analysis method, where the connected domain analysis method is used to find and mark adjacent pixels with the same pixel value in the image. The foreground pixel total number refers to the pixel total number corresponding to the vehicle fascia region in each image in the vehicle fascia injury set, and the background pixel total number is the total number of all pixels excluding the vehicle fascia region in each image in the vehicle fascia injury set.
In this optional embodiment, the ratio of the total number of foreground pixels to the total number of background pixels of each image in the vehicle fascia injury set is used as the initial injury assessment weight corresponding to the image; and normalizing the initial loss assessment weights of all the images in the vehicle rib line injury set to obtain the loss assessment weight of each image in the vehicle rib line injury set.
For example, there are A, B, C, D images in the vehicle rib line injury set, and each image corresponds to an initial loss assessment weight of 0.2, 0.12, 0.16 and 0.32, and then the loss assessment weights obtained by normalizing the initial loss assessment weights of the four images are respectively 0.25, 0.15, 0.2 and 0.4.
In this alternative embodiment, the weighted summation may be performed based on the impairment weight of each image in the vehicle fascia injury set and the corresponding image similarity, so as to obtain the overall similarity between all the images in the vehicle fascia injury set and the vehicle fascia standard image, and the overall similarity is used as the vehicle fascia injury degree of the vehicle fascia at the corresponding point of the vehicle fascia injury set.
For example, there are A, B, C, D images in the vehicle fascia injury set, and the similarity between the images and the vehicle fascia standard image is 0.4, 0.8, 0.3, and 0.5, and the corresponding impairment weights are 0.25, 0.15, 0.2, and 0.4, respectively, so that the final calculated vehicle fascia injury is 0.4x0.25+0.8x0.15+0.3x0.2+0.5x0.4=0.48.
In this alternative embodiment, the historical data of all the vehicle tendon injury degrees corresponding to the vehicle tendons at each point location and the damage assessment line corresponding to each vehicle tendon injury degree may be stored in the database. When the latest obtained vehicle rib line damage degree is subjected to damage assessment, the average value of all the vehicle rib line damage degrees corresponding to the vehicle rib line at the current point position and the average value of the damage assessment amount corresponding to each vehicle rib line damage degree can be used as a standard damage assessment data pair, and the damage assessment amount corresponding to the current vehicle rib line damage degree can be calculated by referring to the standard damage assessment data pair.
For example, there are 4 damage conditions in the vehicle fascia at the vehicle hood position in the history, the damage degree of each vehicle fascia is respectively 0.5, 0.6, 0.2 and 0.3, the corresponding damage limit is 5000, 8000, 3000 and 4000, the average value of the damage degree of the vehicle fascia is calculated to obtain 0.4, the average value of the damage limit is calculated to obtain 5000, 8000, 3000 and 4000 is calculated to obtain 5000, and the corresponding standard damage data pair is (0.4 and 5000). If the current calculated vehicle fascia injury degree is 0.48, the current vehicle fascia injury degree corresponds to an estimated injury limit of 0.48/0.4x5000=6000.
In this optional embodiment, after the current vehicle rib line damage degree and the corresponding damage assessment limit are obtained through each calculation, the standard damage assessment data pair can be updated, so that when the vehicle rib line damage is assessed, a vehicle claim settlement person can refer to the historical data corresponding to all damage assessment limits of the vehicle rib line at the current point location, and more objective and fair judgment can be made on the current vehicle rib line damage to obtain more accurate damage assessment limit.
According to the technical scheme, the obtained vehicle fascia images can be effectively enhanced, the preset target detection network is trained according to the enhanced images to obtain the vehicle fascia detection model, and damaged vehicle fascia can be further subjected to damage assessment according to the vehicle fascia damage set detected by the vehicle fascia detection model, so that the accuracy of detecting the vehicle fascia damage is improved.
Fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present application. The electronic device 1 comprises a memory 12 and a processor 13. The memory 12 is configured to store computer readable instructions and the processor 13 is configured to execute the computer readable instructions stored in the memory to implement the artificial intelligence based vehicle fascia detection method according to any of the embodiments described above.
In an alternative embodiment, the electronic device 1 further comprises a bus, a computer program stored in said memory 12 and executable on said processor 13, such as an artificial intelligence based vehicle fascia detection program.
Fig. 3 shows only an electronic device 1 with a memory 12 and a processor 13, it being understood by a person skilled in the art that the structure shown in fig. 3 does not constitute a limitation of the electronic device 1, and may comprise fewer or more components than shown, or may combine certain components, or a different arrangement of components.
In connection with fig. 1, the memory 12 in the electronic device 1 stores a plurality of computer readable instructions to implement an artificial intelligence based vehicle fascia detection method, the processor 13 being executable to implement:
collecting vehicle rib line images at different positions to obtain various types of vehicle rib line image sets;
Performing image enhancement on all images in the vehicle rib image set to obtain a rib image enhancement set;
training a preset target detection network based on the reinforcement line image enhancement set to obtain a vehicle reinforcement line detection model;
detecting the vehicle fascia image set based on the vehicle fascia detection model to obtain a vehicle fascia damage set, wherein the vehicle fascia image set corresponds to the vehicle fascia damage set one by one;
and determining damage to the damaged vehicle fascia based on the vehicle fascia damage set and a preset vehicle fascia standard image.
Specifically, the specific implementation method of the above instructions by the processor 13 may refer to the description of the relevant steps in the corresponding embodiment of fig. 1, which is not repeated herein.
It will be appreciated by those skilled in the art that the schematic diagram is merely an example of the electronic device 1 and does not constitute a limitation of the electronic device 1, the electronic device 1 may be a bus type structure, a star type structure, the electronic device 1 may further comprise more or less other hardware or software than illustrated, or a different arrangement of components, e.g. the electronic device 1 may further comprise an input-output device, a network access device, etc.
It should be noted that the electronic device 1 is only used as an example, and other electronic products that may be present in the present application or may be present in the future are also included in the scope of the present application and are incorporated herein by reference.
The memory 12 includes at least one type of readable storage medium, which may be non-volatile or volatile. The readable storage medium includes flash memory, a removable hard disk, a multimedia card, a card type memory (e.g., SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. The memory 12 may in some embodiments be an internal storage unit of the electronic device 1, such as a mobile hard disk of the electronic device 1. The memory 12 may in other embodiments also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which are provided on the electronic device 1. The memory 12 may be used not only for storing application software installed in the electronic device 1 and various types of data, such as codes of an artificial intelligence-based vehicle fascia detection program, etc., but also for temporarily storing data that has been output or is to be output.
The processor 13 may be comprised of integrated circuits in some embodiments, for example, a single packaged integrated circuit, or may be comprised of multiple integrated circuits packaged with the same or different functions, including one or more central processing units (Central Processing unit, CPU), microprocessors, digital processing chips, graphics processors, a combination of various control chips, and the like. The processor 13 is a Control Unit (Control Unit) of the electronic device 1, connects the respective components of the entire electronic device 1 using various interfaces and lines, executes various functions of the electronic device 1 and processes data by running or executing programs or modules stored in the memory 12 (for example, executing an artificial intelligence-based vehicle fascia detection program or the like), and calling data stored in the memory 12.
The processor 13 executes the operating system of the electronic device 1 and various types of applications installed. The processor 13 executes the application program to implement the steps described above in various embodiments of the artificial intelligence based vehicle fascia detection method, such as the steps shown in fig. 1.
Illustratively, the computer program may be partitioned into one or more modules/units that are stored in the memory 12 and executed by the processor 13 to complete the present application. The one or more modules/units may be a series of computer readable instruction segments capable of performing the specified functions, which instruction segments describe the execution of the computer program in the electronic device 1. For example, the computer program may be divided into an acquisition unit 110, an enhancement unit 111, a training unit 112, a detection unit 113, a loss assessment unit 114.
The integrated units implemented in the form of software functional modules described above may be stored in a computer readable storage medium. The software functional modules described above are stored in a storage medium and include instructions for causing a computer device (which may be a personal computer, a computer device, or a network device, etc.) or processor (processor) to perform portions of the artificial intelligence-based vehicle fascia detection methods described in various embodiments of the application.
The integrated modules/units of the electronic device 1 may be stored in a computer readable storage medium if implemented in the form of software functional units and sold or used as a stand alone product. Based on such understanding, the present application may implement all or part of the flow of the method of the above embodiment, or may be implemented by instructing the relevant hardware device by a computer program, where the computer program may be stored in a computer readable storage medium, and the computer program may implement the steps of each method embodiment described above when executed by a processor.
Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM), a random access Memory, other memories, and the like.
Further, the computer-readable storage medium may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function, and the like; the storage data area may store data created from the use of blockchain nodes, and the like.
The blockchain referred to in the application is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm and the like. The Blockchain (Blockchain), which is essentially a decentralised database, is a string of data blocks that are generated by cryptographic means in association, each data block containing a batch of information of network transactions for verifying the validity of the information (anti-counterfeiting) and generating the next block. The blockchain may include a blockchain underlying platform, a platform product services layer, an application services layer, and the like.
The bus may be a peripheral component interconnect standard (Peripheral Component Interconnect, PCI) bus or an extended industry standard architecture (Extended Industry Standard Architecture, EISA) bus, among others. The bus may be classified as an address bus, a data bus, a control bus, etc. For ease of illustration, only one arrow is shown in FIG. 3, but only one bus or one type of bus is not shown. The bus is arranged to enable a connection communication between the memory 12 and at least one processor 13 or the like.
The embodiment of the application further provides a computer readable storage medium (not shown), in which computer readable instructions are stored, and the computer readable instructions are executed by a processor in an electronic device to implement the vehicle fascia detection method based on artificial intelligence according to any one of the embodiments.
In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be other manners of division when actually implemented.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.
Furthermore, it is evident that the word "comprising" does not exclude other elements or steps, and that the singular does not exclude a plurality. Several of the elements or devices described in the specification may be embodied by one and the same item of software or hardware. The terms first, second, etc. are used to denote a name, but not any particular order.
Finally, it should be noted that the above embodiments are merely for illustrating the technical solution of the present application and not for limiting, and although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present application may be modified or substituted without departing from the spirit and scope of the technical solution of the present application.
Claims (10)
1. An artificial intelligence based vehicle fascia detection method, the method comprising:
collecting vehicle rib line images at different positions to obtain various types of vehicle rib line image sets;
performing image enhancement on all images in the vehicle rib image set to obtain a rib image enhancement set;
training a preset target detection network based on the reinforcement line image enhancement set to obtain a vehicle reinforcement line detection model;
detecting the vehicle fascia image set based on the vehicle fascia detection model to obtain a vehicle fascia damage set, wherein the vehicle fascia image set corresponds to the vehicle fascia damage set one by one;
and determining damage to the damaged vehicle fascia based on the vehicle fascia damage set and a preset vehicle fascia standard image.
2. The artificial intelligence based vehicle fascia detection method according to claim 1, wherein the acquiring vehicle fascia images at different locations to obtain a plurality of categories of vehicle fascia image sets comprises:
Acquiring vehicle fascia according to preset points to obtain vehicle fascia initial image sets of various categories, wherein the preset points correspond to the vehicle fascia initial image sets one by one;
and carrying out edge detection on all images in the vehicle fascia initial image set to obtain a plurality of types of vehicle fascia image sets.
3. The artificial intelligence based vehicle fascia detection method according to claim 1, wherein said image enhancing all images in said vehicle fascia image set to obtain a fascia image enhancement set comprises:
performing morphological operation on all images in the vehicle rib image set to obtain a rib image set;
and carrying out data enhancement on all images in the tendon image set to obtain a tendon image enhancement set.
4. The artificial intelligence based vehicle fascia detection method according to claim 1, wherein training a preset target detection network based on the fascia image enhancement set to obtain a vehicle fascia detection model comprises:
setting labels for all images in the reinforcement line image enhancement set according to a preset mode to obtain a reinforcement line image label set, wherein the reinforcement line image enhancement set corresponds to the reinforcement line image label set one by one;
Training a preset target detection network based on the fascia image enhancement set and the fascia image tag set to obtain a vehicle fascia detection model.
5. The artificial intelligence based vehicle fascia detection method according to claim 1, wherein the detecting the vehicle fascia image set based on the vehicle fascia detection model to obtain a vehicle fascia injury set, the vehicle fascia image set and the vehicle fascia injury set being in one-to-one correspondence, comprises:
inputting all images in the vehicle fascia image set into the vehicle fascia detection model to obtain vehicle fascia damage pictures comprising a plurality of vehicle fascia damage types;
and taking the obtained vehicle fascia injury pictures comprising various vehicle fascia injury types as a vehicle fascia injury set, wherein the vehicle fascia image set corresponds to the vehicle fascia injury set one by one.
6. The artificial intelligence based vehicle fascia detection method according to claim 1, wherein said determining damage to a damaged vehicle fascia based on the set of vehicle fascia damage and a predetermined vehicle fascia standard image comprises:
calculating the image similarity between each image in the vehicle rib line damage set and a preset vehicle rib line standard image;
Calculating the damage assessment weight of each image in the vehicle rib line damage set;
carrying out weighted summation based on the loss assessment weight and the image similarity to obtain vehicle rib line damage degree, wherein the vehicle rib line damage degree corresponds to the vehicle rib line damage set one by one;
and determining damage to the damaged vehicle fascia based on the damage degree of the vehicle fascia.
7. The artificial intelligence based vehicle fascia detection method according to claim 6, wherein said calculating the impairment weights for each image in the set of vehicle fascia impairment comprises:
counting the total number of foreground pixels and the total number of background pixels of each image in the vehicle fascia injury set according to a connected domain analysis method;
taking the ratio of the total number of foreground pixels to the total number of background pixels of each image in the vehicle rib line damage set as the initial damage assessment weight corresponding to the image;
normalizing the initial loss assessment weights of all the images in the vehicle rib line injury set to obtain the loss assessment weight of each image in the vehicle rib line injury set.
8. An artificial intelligence based vehicle fascia detection apparatus, the apparatus comprising:
the acquisition unit is used for acquiring vehicle rib line images at different positions to obtain various types of vehicle rib line image sets;
The enhancement unit is used for carrying out image enhancement on all images in the vehicle rib image set to obtain a rib image enhancement set;
the training unit is used for training a preset target detection network based on the reinforcement line image enhancement set to obtain a vehicle reinforcement line detection model;
the detection unit is used for detecting the vehicle fascia image set based on the vehicle fascia detection model to obtain a vehicle fascia damage set, and the vehicle fascia image set corresponds to the vehicle fascia damage set one by one;
and the damage assessment unit is used for assessing damaged vehicle fascia based on the vehicle fascia damage set and a preset vehicle fascia standard image.
9. An electronic device, the electronic device comprising:
a memory storing computer readable instructions; and
A processor executing computer readable instructions stored in the memory to implement the artificial intelligence based vehicle fascia detection method of any of claims 1 to 7.
10. A computer readable storage medium having computer readable instructions stored thereon, which when executed by a processor, implement the artificial intelligence based vehicle fascia detection method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311685879.3A CN117611569A (en) | 2023-12-08 | 2023-12-08 | Vehicle fascia detection method, device, equipment and medium based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311685879.3A CN117611569A (en) | 2023-12-08 | 2023-12-08 | Vehicle fascia detection method, device, equipment and medium based on artificial intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117611569A true CN117611569A (en) | 2024-02-27 |
Family
ID=89956128
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311685879.3A Pending CN117611569A (en) | 2023-12-08 | 2023-12-08 | Vehicle fascia detection method, device, equipment and medium based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117611569A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117893413A (en) * | 2024-03-15 | 2024-04-16 | 博创联动科技股份有限公司 | Vehicle-mounted terminal man-machine interaction method based on image enhancement |
-
2023
- 2023-12-08 CN CN202311685879.3A patent/CN117611569A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117893413A (en) * | 2024-03-15 | 2024-04-16 | 博创联动科技股份有限公司 | Vehicle-mounted terminal man-machine interaction method based on image enhancement |
| CN117893413B (en) * | 2024-03-15 | 2024-06-11 | 博创联动科技股份有限公司 | Vehicle-mounted terminal man-machine interaction method based on image enhancement |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111695609B (en) | Target damage degree judging method and device, electronic equipment and storage medium | |
| CN113705462B (en) | Face recognition method, device, electronic equipment and computer readable storage medium | |
| CN117611569A (en) | Vehicle fascia detection method, device, equipment and medium based on artificial intelligence | |
| CN113222063B (en) | Express carton garbage classification method, device, equipment and medium | |
| CN115063632A (en) | Vehicle damage identification method, device, equipment and medium based on artificial intelligence | |
| CN109800745B (en) | Bill detection method, device, terminal and computer readable storage medium | |
| CN115083008A (en) | Moving object detection method, device, equipment and storage medium | |
| CN113191189A (en) | Face living body detection method, terminal device and computer readable storage medium | |
| CN114998282B (en) | Image detection method, device, electronic equipment and storage medium | |
| CN116363365A (en) | Image segmentation method based on semi-supervised learning and related equipment | |
| CN116452802A (en) | Vehicle loss detection method, device, equipment and storage medium | |
| CN115984786A (en) | Vehicle damage detection method and device, terminal and storage medium | |
| CN115131564A (en) | Vehicle component damage detection method based on artificial intelligence and related equipment | |
| CN111583215A (en) | Intelligent damage assessment method and device for damage image, electronic equipment and storage medium | |
| CN112541899B (en) | Incomplete detection method and device of certificate, electronic equipment and computer storage medium | |
| CN115731179A (en) | Track component detection method, terminal and storage medium | |
| CN113239738B (en) | Image blurring detection method and blurring detection device | |
| CN117173075A (en) | Medical image detection method and related equipment | |
| CN114943908B (en) | Vehicle body damage evidence obtaining method, device, equipment and medium based on artificial intelligence | |
| CN114972883B (en) | Target detection sample generation method based on artificial intelligence and related equipment | |
| CN113221907A (en) | Vehicle part segmentation method, device, equipment and storage medium | |
| CN114943865B (en) | Target detection sample optimization method based on artificial intelligence and related equipment | |
| CN114972761B (en) | Vehicle part segmentation method based on artificial intelligence and related equipment | |
| CN117689479A (en) | Vehicle claim recording method, device, equipment and medium based on artificial intelligence | |
| CN114842205B (en) | Vehicle loss detection method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |