CN111612727A - Method for mapping pressure footprint image by optical footprint image - Google Patents
Method for mapping pressure footprint image by optical footprint image Download PDFInfo
- Publication number
- CN111612727A CN111612727A CN202010395588.0A CN202010395588A CN111612727A CN 111612727 A CN111612727 A CN 111612727A CN 202010395588 A CN202010395588 A CN 202010395588A CN 111612727 A CN111612727 A CN 111612727A
- Authority
- CN
- China
- Prior art keywords
- footprint image
- image
- pressure
- optical
- mapping
- 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.)
- Granted
Links
- 230000003287 optical effect Effects 0.000 title claims abstract description 53
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000013507 mapping Methods 0.000 title claims abstract description 22
- 238000012549 training Methods 0.000 claims abstract description 12
- 238000001914 filtration Methods 0.000 claims abstract description 10
- 230000036544 posture Effects 0.000 claims description 9
- 210000004744 fore-foot Anatomy 0.000 claims description 3
- 230000000877 morphologic effect Effects 0.000 abstract description 2
- 238000011160 research Methods 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 230000017105 transposition Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- 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]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method for mapping an optical footprint image to a pressure footprint image, which comprises the following steps: s1, collecting an optical footprint image; s2, collecting a pressure footprint image; s3, carrying out filtering and denoising on the pressure footprint image collected in S2; s4, combining the optical footprint image in S1 and the denoised pressure footprint image in S3 into an image as a training image; s5, importing a plurality of training images into a computer, and obtaining an optimal model through continuous game of a generator and a discriminator by utilizing a Pix2Pix algorithm; and S6, collecting a single optical footprint image, importing the single optical footprint image into the optimal model obtained in S5, and mapping the pressure footprint image. The method for mapping the optical footprint image to the pressure footprint image can realize mapping of the optical footprint image to the pressure footprint image, so that a reliable pressure footprint image of a target individual is provided for police and scientific research workers for analyzing morphological characteristics of the target individual.
Description
Technical Field
The invention relates to the technical field of image processing and biological feature recognition, in particular to a method for mapping an optical footprint image to a pressure footprint image.
Background
The Pix2Pix can realize the direct conversion of the matched images in the two fields, and the obtained result is clearer; the U-net network is mainly thought of as being derived from FCN, adopts a full convolution network to classify images pixel by pixel, can achieve good effect in the field of image segmentation, is named after the network structure is similar to a U shape, and consists of a convolution compression layer at the left end and a transposition convolution amplification layer at the right end; the PatchGAN discriminator is one of discrimination models, and intuitively, the PatchGAN discriminator is completely composed of convolution layers, and finally outputs a matrix of n x n, and finally takes the average value of the output matrix as the output of True/False. In fact, each output in the output matrix represents a field of view in the original, corresponding to a slice (patch) of the original.
At present, the pressure footprint image is obtained by a person to be collected standing or walking on a pressure plate, manual operation is needed, the machine cost is high, and regular maintenance is needed to prevent the damage of the pressure collector from influencing the normal collection of the pressure footprint. And for crime scenes, it is more impossible to obtain a pressure footprint image of a suspect in this way.
Disclosure of Invention
Technical problem to be solved
Aiming at the defects of the prior art, the invention provides a method for mapping an optical footprint image to a pressure footprint image, which can realize the mapping of the optical footprint image to the pressure footprint image, thereby providing a reliable pressure footprint image of a target individual for police and scientific research workers for analyzing the morphological characteristics of the target individual.
(II) technical scheme
In order to achieve the purpose, the invention is realized by the following technical scheme: a method of mapping an optical footprint image to a pressure footprint image, comprising the steps of:
s1, collecting an optical footprint image;
s2, collecting a pressure footprint image;
s3, carrying out filtering and denoising on the pressure footprint image collected in S2;
s4, combining the optical footprint image in S1 and the denoised pressure footprint image in S3 into an image as a training image;
s5, importing a plurality of training images into a computer, and obtaining an optimal model through continuous game of a generator and a discriminator by utilizing a Pix2Pix algorithm;
and S6, collecting a single optical footprint image, importing the single optical footprint image into the optimal model obtained in S5, and mapping the pressure footprint image.
Preferably, in step S1, an optical device is used to collect the optical footprint, and the user walks in different postures.
Preferably, in step S2, the footprint collector is used to collect the pressure footprint image information, and the walking posture is the same as the walking posture on the optical device.
Preferably, the denoising method in step S3 includes: and constructing a 3 multiplied by 3 filtering window by taking the current pressure point as a center, counting the number of pressure values in the filtering window, and if the number of the pressure values is less than a certain threshold value, rejecting the pressure point.
Preferably, in step S5, the generator adopts a U-net network structure, and the arbiter adopts a PatchGAN network structure.
Preferably, the number of training images in step S5 is not less than 6000.
Preferably, in the step S6, the single captured optical footprint image is divided into two images, namely, a forefoot image and a heel image, and then the two images are imported into the optimal model.
(III) advantageous effects
The invention provides a method for mapping an optical footprint image to a pressure footprint image, which has the following beneficial effects: the optical footprint image can be mapped into the pressure footprint image, so that a foundation is laid for the subsequent extraction of pressure characteristics, and meanwhile, instrument equipment and necessary manual operation are omitted.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is an optical footprint map collected by the present invention;
FIG. 3 is a pressure footprint map collected by the present invention;
FIG. 4 is a merged training image of the present invention;
FIG. 5 is a front sole portion of a single optical footprint image of the present invention;
FIG. 6 is a single optical footprint image back heel portion of the present invention;
FIG. 7 is a pressure footprint image mapped by the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1-7, the present invention provides a technical solution: a method of mapping an optical footprint image to a pressure footprint image, comprising the steps of:
s1, collecting an optical footprint image;
s2, collecting a pressure footprint image;
s3, carrying out filtering and denoising on the pressure footprint image collected in S2;
s4, combining the optical footprint image in S1 and the denoised pressure footprint image in S3 into an image as a training image;
s5, importing a plurality of training images into a computer, and obtaining an optimal model through continuous game of a generator and a discriminator by utilizing a Pix2Pix algorithm;
and S6, collecting a single optical footprint image, importing the single optical footprint image into the optimal model obtained in S5, and mapping the pressure footprint image.
In the step S1, optical equipment is used to collect the optical footprints, and the user walks according to different postures, such as slow walking, fast walking, normal walking or squatting.
In the step S2, a footprint collector is used to collect the pressure footprint image information, and the walking posture is the same as the walking posture on the optical device.
The denoising method in the step S3 includes: and constructing a 3 multiplied by 3 filtering window by taking the current pressure point as a center, counting the number of pressure values in the filtering window, and if the number of the pressure values is smaller than a certain threshold value, rejecting the pressure point, so as to reduce noise interference and avoid data clutter accumulation.
The generator in step S5 adopts a U-net network architecture in which the input will go through a series of layers that are progressively downsampled until it becomes the bottleneck layer, at which point the process will be reversed. Such networks require all information flows to go through all layers, including bottlenecks. For many image translation problems, a large amount of low-level information is shared between input and output, and it is therefore desirable to have this information directly across the network. Specifically, we add a layer jump connection between the ith layer and the n-ith layer, where n is the total number of layers. Each layer jump connection only connects all channels of the ith layer with all channels of the (n-i) th layer; the discriminator adopts a PatchGAN network structure, and the L2 loss and the L1 loss can generate fuzzy results in terms of image generation, namely the L1 and the L2 cannot well restore high-frequency parts (edges and the like in an image) of the image, but can well restore low-frequency parts (color blocks in the image). For the problem in this case, an entirely new framework is not needed to map the correctness of the image at low frequencies, and the L1 penalty is already in place. This forces the GAN discriminator to be limited to simulating high frequency structures, relying on the term L1 to recover the accuracy of low frequency, and in order to build a high frequency model, it is sufficient to focus on the structure of local image blocks, PatchGAN, which penalizes only the patch-scale structure and tries to classify whether each nxn block in the image is true or false. We run this discriminator convolved over the image, averaging all responses to provide the final output of D. We demonstrate that N can be much smaller than the overall size of the image and still produce high quality results.
In the step S5, the training images are not less than 6000 images, and the generators and the discriminators continuously play games to continuously update the optimization model to finally obtain a stable pix2pix model with the optical footprint mapped to the pressure footprint, so that any one optical footprint image can be mapped to the pressure footprint image in real time, and the accuracy is ensured.
In the step S6, the collected single optical footprint image is divided into two images of the forefoot and the heel, and then the two images are imported into the optimal model, and in order to satisfy the image in the form of fig. 4, the optical footprint image is divided into two images.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. The term "comprising", without further limitation, means that the element so defined is not excluded from the group consisting of additional identical elements in the process, method, article, or apparatus that comprises the element.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (7)
1. A method of mapping an optical footprint image to a pressure footprint image, characterized by: the method comprises the following steps:
s1, collecting an optical footprint image;
s2, collecting a pressure footprint image;
s3, carrying out filtering and denoising on the pressure footprint image collected in S2;
s4, combining the optical footprint image in S1 and the denoised pressure footprint image in S3 into an image as a training image;
s5, importing a plurality of training images into a computer, and obtaining an optimal model through continuous game of a generator and a discriminator by utilizing a Pix2Pix algorithm;
and S6, collecting a single optical footprint image, importing the single optical footprint image into the optimal model obtained in S5, and mapping the pressure footprint image.
2. The method of mapping an optical footprint image to a pressure footprint image of claim 1, in which: in step S1, an optical device is used to collect the optical footprints, and the walking device walks in different postures.
3. The method of mapping an optical footprint image to a pressure footprint image of claim 1, in which: in the step S2, a footprint collector is used to collect the pressure footprint image information, and the walking posture is the same as the walking posture on the optical device.
4. The method of mapping an optical footprint image to a pressure footprint image of claim 1, in which: the denoising method in the step S3 includes: and constructing a 3 multiplied by 3 filtering window by taking the current pressure point as a center, counting the number of pressure values in the filtering window, and if the number of the pressure values is less than a certain threshold value, rejecting the pressure point.
5. The method of mapping an optical footprint image to a pressure footprint image of claim 1, in which: in the step S5, the generator adopts a U-net network structure, and the arbiter adopts a PatchGAN network structure.
6. The method of mapping an optical footprint image to a pressure footprint image of claim 1, in which: the number of training images in step S5 is not less than 6000.
7. The method of mapping an optical footprint image to a pressure footprint image of claim 1, in which: in the step S6, the single captured optical footprint image is divided into two images, i.e., a forefoot image and a heel image, and then the two images are imported into the optimal model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010395588.0A CN111612727B (en) | 2020-05-12 | 2020-05-12 | Method for mapping pressure footprint image by using optical footprint image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010395588.0A CN111612727B (en) | 2020-05-12 | 2020-05-12 | Method for mapping pressure footprint image by using optical footprint image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111612727A true CN111612727A (en) | 2020-09-01 |
| CN111612727B CN111612727B (en) | 2023-09-12 |
Family
ID=72200283
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010395588.0A Active CN111612727B (en) | 2020-05-12 | 2020-05-12 | Method for mapping pressure footprint image by using optical footprint image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111612727B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150254821A1 (en) * | 2012-12-26 | 2015-09-10 | Dailian Everspry Sci & Tech Co., Ltd. | Processing method of pressure-sensitive light and shadow imaging system and formed footprint image |
| US20160157725A1 (en) * | 2014-12-08 | 2016-06-09 | Luis Daniel Munoz | Device, system and methods for assessing tissue structures, pathology, and healing |
| CN109325546A (en) * | 2018-10-19 | 2019-02-12 | 大连海事大学 | A kind of combination footwork feature at time footprint recognition method |
-
2020
- 2020-05-12 CN CN202010395588.0A patent/CN111612727B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150254821A1 (en) * | 2012-12-26 | 2015-09-10 | Dailian Everspry Sci & Tech Co., Ltd. | Processing method of pressure-sensitive light and shadow imaging system and formed footprint image |
| US20160157725A1 (en) * | 2014-12-08 | 2016-06-09 | Luis Daniel Munoz | Device, system and methods for assessing tissue structures, pathology, and healing |
| CN109325546A (en) * | 2018-10-19 | 2019-02-12 | 大连海事大学 | A kind of combination footwork feature at time footprint recognition method |
Non-Patent Citations (1)
| Title |
|---|
| 施明智;: "三维激光扫描足迹识别技术应用研究" * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111612727B (en) | 2023-09-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Tang et al. | Pixel convolutional neural network for multi-focus image fusion | |
| CN109886090B (en) | Video pedestrian re-identification method based on multi-time scale convolutional neural network | |
| CN104408429B (en) | A kind of video represents frame extracting method and device | |
| CN110334779B (en) | Multi-focus image fusion method based on PSPNet detail extraction | |
| CN110135319A (en) | A kind of anomaly detection method and its system | |
| Zhou et al. | FSAD-Net: feedback spatial attention dehazing network | |
| CN112990077B (en) | Face action unit identification method and device based on joint learning and optical flow estimation | |
| CN105976383A (en) | Power transmission equipment fault diagnosis method based on limit learning machine image recognition | |
| CN112581438B (en) | Slice image recognition method and device, storage medium and electronic equipment | |
| CN114359727B (en) | Tea disease identification method and system based on lightweight optimization Yolo v4 | |
| CN112434608B (en) | Human behavior identification method and system based on double-current combined network | |
| CN107633226A (en) | A kind of human action Tracking Recognition method and system | |
| CN109671055B (en) | Pulmonary nodule detection method and device | |
| CN111723773A (en) | Remnant detection method, device, electronic equipment and readable storage medium | |
| CN104616026A (en) | Monitor scene type identification method for intelligent video monitor | |
| CN110188694A (en) | It is a kind of that shoeprints recognition sequence method is worn based on pressure characteristic | |
| CN103700109B (en) | SAR image change detection based on multiple-objection optimization MOEA/D and fuzzy clustering | |
| CN114187214A (en) | Infrared and visible light image fusion system and method | |
| WO2022183805A1 (en) | Video classification method, apparatus, and device | |
| CN109784230A (en) | A kind of facial video image quality optimization method, system and equipment | |
| CN112308087A (en) | Integrated imaging identification system and method based on dynamic vision sensor | |
| CN111192206A (en) | Method for improving image definition | |
| CN114372962A (en) | Laparoscopic surgery stage identification method and system based on double-particle time convolution | |
| CN114359167A (en) | Insulator defect detection method based on lightweight YOLOv4 in complex scene | |
| CN111612727A (en) | Method for mapping pressure footprint image by optical footprint image |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |