CN112016845A - DNN and CIM based regional economic benefit evaluation method and system - Google Patents
DNN and CIM based regional economic benefit evaluation method and system Download PDFInfo
- Publication number
- CN112016845A CN112016845A CN202010924586.6A CN202010924586A CN112016845A CN 112016845 A CN112016845 A CN 112016845A CN 202010924586 A CN202010924586 A CN 202010924586A CN 112016845 A CN112016845 A CN 112016845A
- Authority
- CN
- China
- Prior art keywords
- garbage
- overflow
- trash
- rgb
- economic benefit
- 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.)
- Withdrawn
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 31
- 239000010813 municipal solid waste Substances 0.000 claims abstract description 209
- 238000001514 detection method Methods 0.000 claims abstract description 51
- 238000000034 method Methods 0.000 claims abstract description 26
- 238000012545 processing Methods 0.000 claims abstract description 26
- 238000005520 cutting process Methods 0.000 claims abstract description 7
- 238000000605 extraction Methods 0.000 claims description 9
- 238000003860 storage Methods 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 description 4
- 230000000007 visual effect Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/067—Enterprise or organisation modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Processing Of Solid Wastes (AREA)
Abstract
The invention provides a DNN and CIM-based regional economic benefit evaluation method and system, and specifically the implementation process of the invention is as follows: firstly, acquiring an RGB (red, green and blue) image and a depth image of a garbage can, processing the RGB image by using a garbage can detection network to obtain a boundary frame of the garbage can, cutting the RGB image and the depth image by using the boundary frame to obtain an RGB-D image, and sending the RGB-D image into an overflow degree detection network to obtain overflow degree data of the garbage can; and processing the overflow degree data of each garbage can of different types to obtain the overflow times of the garbage cans of different types, and evaluating the economic benefit of each area according to the overflow times and the preset weight parameters to obtain the corresponding economic benefit evaluation value. The invention does not need to weigh the garbage in a garbage disposal plant, has simple execution process and accurate evaluation result.
Description
Technical Field
The invention relates to the field of artificial intelligence and economic benefit evaluation, in particular to a DNN and CIM based regional economic benefit evaluation method and system.
Background
The existing assessment aiming at the economic benefit of urban regional garbage reduction treatment has no qualitative analysis, the economic benefit is generally quantified by the weight of garbage directly in a garbage treatment plant, and the method cannot consider the garbage amount of each region, so that the economic benefit of each region cannot be assessed.
Disclosure of Invention
In order to solve the above problems, the present invention provides a method and a system for evaluating economic benefits of a region based on DNN and CIM, and more particularly, a method for evaluating economic benefits of a region based on DNN and CIM, the method comprising:
step one, constructing a city information model CIM, and carrying out region division in the CIM;
acquiring an RGB (red, green and blue) image and a depth image of the garbage can in each region by using an RGB-D (red, green and blue) camera in each region;
processing the RGB map by using a garbage can detection network, and outputting coordinates of a garbage can central point and the length and width of a garbage can boundary frame; the central points of the garbage cans of different categories correspond to different channels, and the categories comprise recoverable garbage, harmful garbage, wet garbage and dry garbage;
thirdly, cutting the RGB image and the depth image of the garbage can by using the bounding box of the garbage can to obtain an RGB-D image; sending the RGB-D image into an overflow degree detection network to obtain overflow degree data of the garbage can;
step four, presetting a statistical period, obtaining the overflow degree data of each garbage can according to a fixed time interval in the statistical period, processing the overflow degree data of each garbage can obtained in the statistical period to obtain a change curve of the overflow degree data, and obtaining the overflow times of different types of garbage cans according to the change curve;
fifthly, according to the economic value obtained by processing each type of garbage by the garbage processing station, giving a weight coefficient to each type of garbage, and evaluating the economic benefit of each area in the statistical period according to the weight coefficient and the overflow times to obtain the economic benefit evaluation value of each area;
and step six, storing the obtained economic benefit evaluation value of each area into the CIM.
The invention also provides a DNN and CIM-based regional economic benefit evaluation system, which comprises:
the CIM module is used for constructing a CIM (city information model) and performing region division in the CIM;
the garbage bin detection module is used for acquiring an RGB (red, green and blue) image and a depth image of a garbage bin in each region by using the RGB-D camera in each region; processing the RGB map by using a garbage can detection network, and outputting coordinates of a garbage can central point and the length and width of a garbage can boundary frame; the central points of the garbage cans of different categories correspond to different channels, and the categories comprise recoverable garbage, harmful garbage, wet garbage and dry garbage;
the overflow degree detection module is used for cutting the RGB image and the depth image of the garbage can by using the bounding box of the garbage can obtained by the garbage can detection module to obtain an RGB-D image; sending the RGB-D image into an overflow degree detection network to obtain overflow degree data of the garbage can;
the times counting module is used for processing the overflow degree data of each garbage can obtained in the counting period to obtain a change curve of the overflow degree data in time, and obtaining the overflow times of the garbage cans of different types according to the change curve; the statistical period is preset, and the overflow degree data of each garbage can is obtained by an overflow degree detection module at fixed time intervals in the statistical period; the number of wave crests in the change curve represents the overflowing frequency of the garbage can;
the economic benefit evaluation module is used for evaluating the economic benefit of each area in the statistical period according to the weight coefficient and the overflow times to obtain an economic benefit evaluation value of each area; wherein the weight coefficient of each type of garbage is obtained according to the economic value obtained by processing each type of garbage by the garbage processing station;
and the storage module is used for storing the economic benefit evaluation value of each area obtained by the economic benefit evaluation module into the CIM.
The invention has the beneficial effects that:
1. the invention evaluates the regional economic benefit according to the overflowing frequency of the garbage can, does not need to weigh the garbage in a garbage disposal plant, has simple execution process and accurate evaluation result.
2. The invention can obtain the overflow degree of different types of garbage cans, and can also judge the garbage amount in the garbage can, not only judge whether the garbage can overflows.
3. The method can evaluate the economic benefit of one area, and can also evaluate the economic benefit of one city by combining the economic benefit evaluation values of a plurality of areas.
4. The invention combines CIM and Web GIS technology, displays the overflow degree condition of each road garbage can in the city in Web, can reflect the overflow state of each garbage can in time, and is convenient for related personnel to carry out work.
Drawings
FIG. 1 is a flow chart of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, the following further description is provided in conjunction with the embodiments and the accompanying drawings.
The invention aims to detect the urban garbage bin and detect the overflow degree condition of the urban garbage bin, thereby evaluating the economic benefit of an urban area. Specifically, the invention provides a DNN and CIM based regional economic benefit evaluation method and system, the implementation process of the method is shown in figure 1, the positions of garbage cans are determined and classified by cameras, the overflow degree of the garbage cans is detected, the overflow conditions of different types of garbage cans are finally obtained, and then the weighted evaluation of the economic benefits of garbage treatment is carried out according to the garbage types and the overflow times of the day.
Example (b):
constructing a City Information Model (CIM), and performing region division in the CIM; the CIM is an organic complex of a three-dimensional city space model and building information established by superposing BIM information of city buildings, underground facilities and city internet of things information on the basis of city information data, and mainly comprises GIS data and BIM data of city roads, buildings and infrastructure.
In the method, the CIM technology realizes the datamation and informatization of buildings in the city through a three-dimensional model established in the early stage of the buildings, provides geographical position information for the subsequent overflow detection of the urban garbage can, and combines the DNN technology to realize all-weather overflow detection of the urban garbage can and strengthen urban environment management and economic management.
Acquiring images of the garbage can by using RGB-D depth cameras with different visual angles and different visual distances in each region, and acquiring an RGB image and a depth image of the garbage can in each region; the arrangement of the RGB-D camera suggests that the garbage can is shot in the overlooking direction, so that the depth information of the subsequently obtained garbage can image is more practical, namely the depth information in the image can accurately reflect the stacking height of the garbage in the garbage can.
Processing the RGB map by using a garbage can detection network, and outputting coordinates of a garbage can central point and the length and width of a garbage can boundary frame; the central points of the garbage cans of different categories correspond to different channels, and the categories comprise recoverable garbage, harmful garbage, wet garbage and dry garbage.
The garbage can detection network proposes to utilize a Centernet (object as points) target detection network, and compared with an anchor-based target detection network, the garbage can detection network has good simplicity, and the speed and the precision can achieve the real-time effect; the training process of the network is as follows:
(a) acquiring images of the trash can by using cameras with various visual angles and visual distances to construct a training data set; therefore, the image acquisition is more beneficial to improving the generalization capability of the network.
(b) The manufacturing of the label is well known, and is not described herein any more, and it is to be noted that different types of trash cans need different channels, that is, a trash can central point Heatmap is a multi-channel, the type of the trash can is mainly determined according to a municipal trash can facility, in the embodiment, the trash can is divided into 4 types of trash cans according to four types of trash such as recoverable trash, harmful trash, wet trash and dry trash, the central point of each type of trash can corresponds to an individual channel, and the central points of the 4 types of trash cans are respectively in 4 different channels.
(c) The image data and the Heatmap label data in the training dataset need to be normalized, that is, the picture matrix is changed into floating point numbers between [0,1] so that the model can be converged better, and then the processed image data and label data are sent to a garbage can detection network for training so as to predict the position information (the position x of the central point, the y coordinate and the length and width of the bounding box Bbox, specifically:
the garbage bin detects encoder and carries out the feature extraction to the RGB picture of garbage bin, obtains first feature map, and the garbage bin detects the decoder and carries out upsampling and feature extraction to first feature map, and the output is two, firstly the central point Heatmap of garbage bin, secondly the length and width map of garbage bin boundary frame Bbox.
(d) The loss function adopts the weighted sum of the central point prediction loss and the size loss of the garbage can, wherein the mathematical formula of the central point loss is as follows:
wherein alpha and beta are hyper-parameters, which are set by artificial experience, and N is the number of central points of the trash can in the image. Gamma rayxyFor the predicted value of x, y coordinate in Heatmap, yxyIs the value of x, y coordinate in group Truth Heatmap.
The mathematical formula of the size loss of the garbage can is as follows:
n is the number of trash cans in the image, SPkFor the predicted length and width of the bounding box, skThe length and width of the box are enclosed for group Truth.
The overall loss function is:
Total Loss＝CenterLoss+*SizeLoss
the weight, i.e., the weight of SizeLoss, is usually 0.1.
Thus, the central point Heatmap of the trash can and the length and width map of the bounding box Bbox can be obtained, and then post-processing is performed on the obtained central point Heatmap and the length and width map to obtain specific Bbox information.
Cutting the RGB image and the depth image of the garbage can, namely cutting the image collected by the camera according to the central point position and the length and width information of the garbage can to obtain a 4-channel RGB-D image; in the embodiment, the classification of the garbage can is divided into 4 types, so that a 4-channel RGB-D graph is obtained.
Sending the RGB-D into an overflow degree detection network to obtain overflow degree data of the garbage can; the training process of the overflow degree detection network is as follows:
(a) the garbage can image is sampled to a fixed size, and then normalization processing is carried out, namely the image matrix is changed into floating point numbers between [0 and 1] so that the model can be converged better.
(b) And labeling the overflow degree of the garbage can image, wherein the overflow degree is small, half full and full respectively and is represented by index values 0,1 and 2 respectively, and the index values are overflow degree data.
(c) Sending the processed image data and the tag data to an overflow degree detection network for training to obtain the overflow condition of the urban garbage can, wherein the tag data needs to be subjected to one-hot encoding operation; specifically, the method comprises the following steps:
the overflow degree detection encoder performs feature extraction on the RGB-D image to obtain a second feature image, performs leveling operation on the second feature image to obtain a one-dimensional vector, and sends the one-dimensional vector to a full-connection layer for processing to obtain the probability of each overflow degree of the garbage can; wherein the fully-connected layer serves to map features to the sample label space.
(d) The loss function is a cross entropy function.
And performing argmax operation on the output of the full connection layer to obtain specific overflow degree data of the garbage can.
The invention has diversified module structures, and in order to give consideration to the overflow detection speed and the accuracy of the garbage can, the garbage can detection Encoder-Decoder recommends applying EfficientNet B0-B3 series and DLA-34 pre-training networks to extract characteristics, thereby ensuring higher precision and higher speed. The overflow degree detection Encoder suggests applying ShuffleNet and MobileNet pre-training image classification networks to extract features, and the method has high efficiency.
The above is the process of acquiring the overflow degree data, and the following takes an area as an example to describe the method for calculating the economic benefit evaluation value.
Presetting a statistical period, wherein the statistical period can be one day or one week, an implementer can determine the statistical period by himself, acquiring overflow degree data on each garbage can time sequence in the statistical period according to a fixed time interval, processing a plurality of overflow degree data of each garbage can acquired in the statistical period to acquire a change curve of the overflow degree data, and acquiring overflow times of different types of garbage cans according to the change curve; the specific method for acquiring the overflow times comprises the following steps:
firstly, constructing a change curve by taking time as an abscissa and overflow degree data, namely index values, as an ordinate;
for a change curve, based on the mathematical principle, the overflow times, namely the number of wave crests is searched, and the specific method is as follows:
(1) the overflow degree data in the variation curve may be represented as V ═ V1, V2, …, vn ], where V1, V2, …, vn is the index value of the overflow degree, and n is the number of overflow degree data obtained in the statistical period.
(2) Calculate the first order difference vector DiffV of V: diffv (i) ═ V (i +1) -V (i), where i ranges from [1, n-1 ].
(3) And carrying out sign taking function operation on the difference vector: trend ═ sign (DiffV), i.e. DiffV is traversed, if DiffV (i) is greater than 0, Trend (i) takes 1; DiffV (i) equals 0, Trend (i) takes 0; if DiffV (i) is less than 0, Trend (i) takes the value-1.
(4) Traversing the Trend vector from the tail, the following operations are carried out:
if Trend (i) is 0 and Trend (i +1) ≧ 0, Trend (i) is 1, i.e., the value of Trend (i) changes from 0 to 1;
if Trend (i) ═ 0 and Trend (i +1) <0, then Trend (i) ═ 1, i.e., the value of Trend (i) changes from 0 to-1;
(5) performing first-order difference operation on the Trend vector to obtain R ═ diff (Trend), namely R (i) ═ Trend (i +1) -Trend (i), traversing the difference vector R:
if R (i) is equal to-2, the overflow degree data V (i +1) obtained at the (i +1) th time in the change curve is a wave peak value;
similarly, if r (i) is 2, the overflow degree data V (i +1) obtained at the i +1 th time in the change curve is a valley value.
And counting the number of-2 in the difference vector R to obtain the number of wave crests, namely the overflow frequency of the garbage can. Wherein, every garbage bin all corresponds a change curve.
It should be noted that the precondition for acquiring the overflowing frequency of the garbage can based on the variation curve is as follows: when the overflowing degree of the garbage can is overflowing, workers can timely process the overflowing garbage can, namely the index value 2 in the change curve cannot continuously appear.
According to the economic value obtained by processing each type of garbage by the garbage processing station, a weight coefficient which is more than 0 is endowed to each type of garbage, the economic benefit of each area in the statistical period is evaluated according to the weight coefficient and the overflow frequency, and the economic benefit evaluation value of each area is obtained, wherein the calculation formula of the economic benefit evaluation value is as follows:
M＝α*a+β*b+γ*c+*d
wherein, alpha, beta and gamma are weight coefficients of recoverable garbage, harmful garbage, wet garbage and dry garbage respectively. a. b, c and d are the overflow times of the garbage cans of different types respectively.
For the weight coefficients of the garbage, the following constraints are provided to ensure that the dynamic change intervals of the weight coefficients of different garbage are consistent: α + β + γ + ═ 1.
According to the method, the economic benefit evaluation value of each area can be obtained, the obtained economic benefit evaluation value of each area is stored in the CIM, the CIM is visualized by utilizing a Web GIS technology, the overflow degree condition of each road garbage can of the city and the economic benefit evaluation value of each area are displayed in the Web, and the city environment management and the economic management are enhanced.
The above description is intended to provide those skilled in the art with a better understanding of the present invention and is not intended to limit the present invention.
Claims (10)
1. A region economic benefit evaluation method based on DNN and CIM is characterized by comprising the following steps:
step one, constructing a city information model CIM, and carrying out region division in the CIM;
acquiring an RGB (red, green and blue) image and a depth image of the garbage can in each region by using an RGB-D (red, green and blue) camera in each region;
processing the RGB map by using a garbage can detection network, and outputting coordinates of a garbage can central point and the length and width of a garbage can boundary frame; the central points of the garbage cans of different categories correspond to different channels, and the categories comprise recoverable garbage, harmful garbage, wet garbage and dry garbage;
thirdly, cutting the RGB image and the depth image of the garbage can by using the bounding box of the garbage can to obtain an RGB-D image; sending the RGB-D image into an overflow degree detection network to obtain overflow degree data of the garbage can;
step four, presetting a statistical period, obtaining the overflow degree data of each garbage can according to a fixed time interval in the statistical period, processing the overflow degree data of each garbage can obtained in the statistical period to obtain a change curve of the overflow degree data, and obtaining the overflow times of different types of garbage cans according to the change curve;
fifthly, according to the economic value obtained by processing each type of garbage by the garbage processing station, giving a weight coefficient to each type of garbage, and evaluating the economic benefit of each area in the statistical period according to the weight coefficient and the overflow times to obtain the economic benefit evaluation value of each area;
and step six, storing the obtained economic benefit evaluation value of each area into the CIM.
2. The method of claim 1, wherein the trash can detection network comprises a trash can detection encoder and a trash can detection decoder, wherein the trash can detection encoder performs feature extraction on the RGB map of the trash can to obtain a first feature map, and the trash can detection decoder performs upsampling and feature extraction on the first feature map to obtain coordinates of a center point of the trash can and a length and a width of a bounding box of the trash can.
3. The method according to claim 1, wherein the overflow degree detection network comprises an overflow degree detection classification encoder and a full connection layer, wherein the overflow degree detection encoder performs feature extraction on the RGB-D graph to obtain a second feature graph, performs a leveling operation on the second feature graph, and sends the second feature graph to the full connection layer for processing to obtain the probability of each overflow degree of the trash can; the degree of overflow includes small amount, half full, and overflow.
4. The method of claim 1, wherein the number of peaks in the curve represents the number of overfills of the trash can.
5. The method of claim 1, wherein the economic benefit assessment value is calculated by:
M＝α*a+β*b+γ*c+*d
wherein, alpha, beta and gamma are respectively the weight coefficients of recoverable garbage, harmful garbage, wet garbage and dry garbage, and a, b, c and d are respectively the overflow times of different garbage cans.
6. The method according to claim 5, wherein the weight coefficients have constraint conditions to ensure that the dynamic change intervals of the weight coefficients are consistent, specifically: α + β + γ + ═ 1.
7. A DNN and CIM based regional economic benefit assessment system is characterized by comprising:
the CIM module is used for constructing a CIM (city information model) and performing region division in the CIM;
the garbage bin detection module is used for acquiring an RGB (red, green and blue) image and a depth image of a garbage bin in each region by using the RGB-D camera in each region; processing the RGB map by using a garbage can detection network, and outputting coordinates of a garbage can central point and the length and width of a garbage can boundary frame; the central points of the garbage cans of different categories correspond to different channels, and the categories comprise recoverable garbage, harmful garbage, wet garbage and dry garbage;
the overflow degree detection module is used for cutting the RGB image and the depth image of the garbage can by using the bounding box of the garbage can obtained by the garbage can detection module to obtain an RGB-D image; sending the RGB-D image into an overflow degree detection network to obtain overflow degree data of the garbage can;
the times counting module is used for processing the overflow degree data of each garbage can obtained in the counting period to obtain a change curve of the overflow degree data in time, and obtaining the overflow times of the garbage cans of different types according to the change curve; the statistical period is preset, and the overflow degree data of each garbage can is obtained by an overflow degree detection module at fixed time intervals in the statistical period; the number of wave crests in the change curve represents the overflowing frequency of the garbage can;
the economic benefit evaluation module is used for evaluating the economic benefit of each area in the statistical period according to the weight coefficient and the overflow times to obtain an economic benefit evaluation value of each area; wherein the weight coefficient of each type of garbage is obtained according to the economic value obtained by processing each type of garbage by the garbage processing station;
and the storage module is used for storing the economic benefit evaluation value of each area obtained by the economic benefit evaluation module into the CIM.
8. The system of claim 7, wherein the trash can detection network in the trash can detection module comprises a trash can detection encoder and a trash can detection decoder, wherein the trash can detection encoder performs feature extraction on the RGB map of the trash can to obtain a first feature map, and the trash can detection decoder performs upsampling and feature extraction on the first feature map to obtain the coordinates of the center point of the trash can and the length and width of the bounding box of the trash can.
9. The system of claim 7, wherein the overfill detection network in the overfill detection module comprises an overfill detection encoder and a full link layer, wherein the overfill detection encoder performs feature extraction on the RGB-D graph to obtain a second feature graph, and performs leveling operation on the second feature graph and sends the second feature graph to the full link layer for processing to obtain the probability of each overfill degree of the trash can; the degree of overflow includes small amount, half full, and overflow.
10. The system of claim 7, wherein the economic benefit assessment value in the economic benefit assessment module is calculated by:
M＝α*a+β*b+γ*c+*d
wherein, alpha, beta and gamma are respectively the weight coefficients of recoverable garbage, harmful garbage, wet garbage and dry garbage, and a, b, c and d are respectively the overflow times of different garbage cans; and the weight coefficient constraint is α + β + γ + ═ 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010924586.6A CN112016845A (en) | 2020-09-05 | 2020-09-05 | DNN and CIM based regional economic benefit evaluation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010924586.6A CN112016845A (en) | 2020-09-05 | 2020-09-05 | DNN and CIM based regional economic benefit evaluation method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112016845A true CN112016845A (en) | 2020-12-01 |
Family
ID=73516369
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010924586.6A Withdrawn CN112016845A (en) | 2020-09-05 | 2020-09-05 | DNN and CIM based regional economic benefit evaluation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112016845A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113450401A (en) * | 2021-07-19 | 2021-09-28 | 北京航空航天大学杭州创新研究院 | Trash can fullness degree determining method, device and equipment and trash can |
| CN116797104A (en) * | 2023-08-21 | 2023-09-22 | 深圳市洛丁光电有限公司 | Smart city distributed data acquisition processing method and system based on AIOT |
-
2020
- 2020-09-05 CN CN202010924586.6A patent/CN112016845A/en not_active Withdrawn
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113450401A (en) * | 2021-07-19 | 2021-09-28 | 北京航空航天大学杭州创新研究院 | Trash can fullness degree determining method, device and equipment and trash can |
| CN116797104A (en) * | 2023-08-21 | 2023-09-22 | 深圳市洛丁光电有限公司 | Smart city distributed data acquisition processing method and system based on AIOT |
| CN116797104B (en) * | 2023-08-21 | 2023-11-10 | 深圳市洛丁光电有限公司 | Smart city distributed data acquisition processing method and system based on AIOT |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110570651B (en) | Road network traffic situation prediction method and system based on deep learning | |
| CN113487066B (en) | Long-time-sequence freight volume prediction method based on multi-attribute enhanced graph convolution-Informer model | |
| CN110245709B (en) | 3D point cloud data semantic segmentation method based on deep learning and self-attention | |
| CN111814623A (en) | Vehicle lane departure visual detection method based on deep neural network | |
| CN106228808B (en) | City expressway travel time prediction method based on Floating Car space-time grid data | |
| CN112800913B (en) | Pavement damage data space-time analysis method based on multi-source feature fusion | |
| CN112016845A (en) | DNN and CIM based regional economic benefit evaluation method and system | |
| CN109934170B (en) | Mine resource statistical method based on computer vision | |
| CN116720984B (en) | Intelligent electric power auditing method based on satellite map technology | |
| CN115984850A (en) | Lightweight remote sensing image semantic segmentation method based on improved Deeplabv3+ | |
| CN117351702A (en) | Intelligent traffic management method based on adjustment of traffic flow | |
| CN116343053A (en) | Automatic solid waste extraction method based on fusion of optical remote sensing image and SAR remote sensing image | |
| CN115497002A (en) | Multi-scale feature fusion laser radar remote sensing classification method | |
| CN115311617A (en) | Method and system for acquiring passenger flow information of urban rail station area | |
| CN116168246A (en) | Method, device, equipment and medium for identifying waste slag field for railway engineering | |
| CN116597270A (en) | Road damage target detection method based on attention mechanism integrated learning network | |
| CN110849821A (en) | Black and odorous water body remote sensing identification method based on Bayesian theorem | |
| CN111340001B (en) | Traffic jam recognition method based on remote sensing image deep learning | |
| CN115497006B (en) | Urban remote sensing image change depth monitoring method and system based on dynamic mixing strategy | |
| CN116229396B (en) | High-speed pavement disease identification and warning method | |
| CN117078925A (en) | Building rubbish annual output accurate calculation method based on RDSA-deep LabV3+ network | |
| CN114820931B (en) | Virtual reality-based CIM (common information model) visual real-time imaging method for smart city | |
| CN114913475B (en) | Urban gridding management method and system based on GIS and machine vision | |
| CN113421067A (en) | Shared bicycle brand identification and quantity management method and system | |
| CN113239815A (en) | Remote sensing image classification method, device and equipment based on real semantic full-network learning |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20201201 |
|
| WW01 | Invention patent application withdrawn after publication |