CN113610700B - Automatic registration method, device and system for global comprehensive improvement of image spots - Google Patents
Automatic registration method, device and system for global comprehensive improvement of image spots Download PDFInfo
- Publication number
- CN113610700B CN113610700B CN202110838660.7A CN202110838660A CN113610700B CN 113610700 B CN113610700 B CN 113610700B CN 202110838660 A CN202110838660 A CN 202110838660A CN 113610700 B CN113610700 B CN 113610700B
- Authority
- CN
- China
- Prior art keywords
- registration
- sequence
- interior angle
- registered
- spot
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 55
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 claims abstract description 42
- 238000012216 screening Methods 0.000 claims abstract description 32
- 238000004364 calculation method Methods 0.000 claims abstract description 6
- 239000011159 matrix material Substances 0.000 claims description 38
- 238000012795 verification Methods 0.000 claims description 17
- 230000035876 healing Effects 0.000 claims description 8
- 238000010586 diagram Methods 0.000 description 7
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000009418 renovation Methods 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/14—Transformations for image registration, e.g. adjusting or mapping for alignment of images
- G06T3/147—Transformations for image registration, e.g. adjusting or mapping for alignment of images using affine transformations
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an automatic registration method, device and system for a global comprehensive improvement image spot. The device comprises a pattern spot acquisition unit, a registration screening unit and a parameter calculation unit. The system includes a user input module, an automatic registration module, and a result generation module. By applying a Smith-Waterman algorithm and a vacancy weight constant model, registration scoring and fraction backtracking are carried out on the interior angles in a first interior angle sequence and a second interior angle sequence which are registered, after a first registration result with the highest degree of registration and a first reference pattern spot corresponding to the first registration result are obtained through screening, an affine parameter group between the first reference pattern spot and a first pattern spot to be registered is calculated, and therefore registration is completed.
Description
Technical Field
The invention relates to the field of automatic registration of global comprehensive renovation spots, in particular to an automatic registration method, device and system of global comprehensive renovation spots.
Background
The comprehensive land reclamation of the whole area is a system engineering which uses land arrangement, reclamation, development and increase and decrease of urban and rural construction land as platforms according to the target and the purpose determined by the general land utilization plan in a certain area, promotes the comprehensive reclamation of the fields, water, roads, forests and villages, improves the production, living conditions and ecological environment of the rural areas, promotes the large-scale agricultural operation, the centralized population living and the industrial gathering development and promotes the urban and rural integrated process. In the processing of the related history data of the global comprehensive improvement, vector data with different coordinate systems in the same region are often encountered, but the conversion parameters cannot be obtained due to the reason that the conversion parameters of the coordinate conversion of different projection coordinate systems relate to confidentiality or the leaving of business history. Mixing the raw materials in a ratio of 1: 10000 topographic maps are used as a working base map, land units (plots) with basically the same landform and land utilization type and basically consistent water and soil loss type are divided into a class, the class is used as a basic investigation unit, and then the unit is sketched on the topographic map to form a map spot. (the minimum image spot area on the image is not less than 0.5cm, the solid area is 0.5 km; the maximum area is not more than 50cm, the solid area is 50km) image segmentation or vector diagram registration, and the whole image is divided into a plurality of small areas, namely the image spots.
In the prior art, polygonal data are superposed in a view through arcgis software, matched points are searched through human eye observation, and then affine parameters are calculated; or determining the two most matched image spots by calculating the matching degree between the image spots with the same number of top points.
However, the prior art still has the following defects: 1. when manual matching is adopted, the related areas of the image layers are wide, the number of the image layers is large, the number and the shapes of the image spots are changed, and the efficiency of manually finding the suitable alignment points in the peaks of the image spots is low; 2. when the method is calculated through an algorithm, the same number of vertex points is required for a plurality of image spots to be registered, so that the use limitation of the algorithm is large.
Therefore, there is a need for an automatic registration method, apparatus and system for global integrated healing of patches.
Disclosure of Invention
In view of the above-mentioned technical problems, an object of the present invention is to provide an automatic registration method, apparatus and system for global integrated regularization patches, so as to improve the automation and efficiency of the registration process of global integrated regularization patches.
The invention provides an automatic registration method of a global comprehensive improvement image spot, which comprises the following steps: acquiring a first internal angle sequence of each of the plurality of reference image spots and a second internal angle sequence of the first image spot to be registered; wherein a first length m of the first sequence of interior angles is not equal to a second length n of the second sequence of interior angles; calculating and screening to obtain a first registration result with the highest registration degree and a first reference pattern spot corresponding to the first registration result according to the plurality of first interior angle sequences, the second interior angle sequence, the Smith-Wattman algorithm and a vacancy weight constant model; the first registration result comprises a registration vertex pair group; and according to the registration vertex pair group, calculating an affine parameter group between the first reference image spot and the first to-be-registered image spot, thereby completing registration.
In one embodiment, after the screening is calculated to obtain the first registration result with the highest registration degree and the first reference image spot corresponding to the first registration result, the automatic registration method further includes: performing affine on the first reference image spot and the image spot to be registered according to a plurality of registration vertex pairs in the registration vertex pair group, and determining whether the first registration result passes verification according to an affine result; if not, stopping the automatic registration process and sending a registration failure prompt to the user.
In one embodiment, before obtaining the first sequence of internal angles of each of the plurality of reference patches and the second sequence of internal angles of the patch to be aligned, the automatic registration method further includes: acquiring a reference layer and a layer to be registered, and overlapping the reference layer and the layer to be registered; and selecting a first pattern spot to be registered from the layer to be registered, and selecting a plurality of reference pattern spots from the reference layer by taking the first pattern spot to be registered as a center according to a preset spatial distance.
In an embodiment, the calculating and screening to obtain a first registration result with a highest registration degree and a first reference pattern corresponding to the first registration result according to the plurality of first interior angle sequences, the second interior angle sequence, the smith-waterman algorithm, and the vacancy weight invariant model specifically includes: according to a Smith-Waterman algorithm and a vacancy weight constant model, sequentially aiming at each first interior angle sequence, respectively calculating the matching score between each first interior angle in the first interior angle sequence and each second interior angle in the second interior angle sequence, and constructing a registration matrix according to the matching scores; respectively backtracking the registration matrixes to respectively obtain an optimal registration sequence corresponding to each registration matrix, a registration score corresponding to the optimal registration sequence and a registration vertex pair group of the optimal registration sequence and the second interior angle sequence; wherein the registration vertex pair group comprises a plurality of registration interior angle pairs and a corresponding plurality of interior angle correspondences; screening a first optimal matching sequence with the highest degree of matching from the optimal matching sequences according to the number of second internal angles included in each optimal matching sequence and the corresponding matching score; and taking the first optimal matching sequence, the registration vertex pair group and the corresponding first reference image spot as a first registration result.
In an embodiment, the constructing a registration matrix according to the matching score specifically includes: respectively constructing a zero matrix according to the length of each first interior angle sequence and the length of each second interior angle sequence; the plurality of zero matrixes correspond to the plurality of first interior angle sequences one by one; and respectively filling the matching scores of each first internal angle and each second internal angle into corresponding positions of the corresponding zero matrix by taking the serial number of the first internal angle as a column coordinate and the serial number of the second internal angle as a row coordinate aiming at each zero matrix and the corresponding first internal angle sequence, so as to respectively obtain the registration matrix corresponding to each first internal angle sequence.
The invention also provides an automatic registration device of the global comprehensive treatment image spots, which comprises an image spot acquisition unit, a registration screening unit and a parameter calculation unit, wherein the image spot acquisition unit is used for acquiring the respective first interior angle sequences of a plurality of reference image spots and the second interior angle sequences of the image spots to be registered; wherein a first length m of the first sequence of interior angles is not equal to a second length n of the second sequence of interior angles; the registration screening unit is used for calculating and screening to obtain a first registration result with the highest registration degree and a first reference pattern spot corresponding to the first registration result according to the plurality of first interior angle sequences, the second interior angle sequence, the Smith-Wattman algorithm and a vacancy weight constant model; the first registration result comprises the number of matched vertexes and a registration vertex pair group; and the parameter calculating unit is used for calculating an affine parameter group between the first reference pattern spot and the pattern spot to be registered according to the registration vertex pair group, so that registration is completed.
In one embodiment, the automatic registration apparatus further comprises a registration verification unit for: performing affine on the first reference image spot and the image spot to be registered according to a plurality of registration vertex pairs in the registration vertex pair group, and determining whether the first registration result passes verification according to an affine result; if not, stopping the automatic registration process and sending a registration failure prompt to the user.
In one embodiment, the automatic registration apparatus further comprises an object selection unit for: acquiring a reference layer and a layer to be registered, and overlapping the reference layer and the layer to be registered; and selecting a first pattern spot to be registered from the layer to be registered, and selecting a plurality of reference pattern spots from the reference layer by taking the first pattern spot to be registered as a center according to a preset spatial distance.
In one embodiment, the registration screening unit is further configured to: according to a Smith-Waterman algorithm and a vacancy weight constant model, sequentially aiming at each first interior angle sequence, respectively calculating the matching score between each first interior angle in the first interior angle sequence and each second interior angle in the second interior angle sequence, and constructing a registration matrix according to the matching scores; respectively backtracking the registration matrixes to respectively obtain an optimal registration sequence corresponding to each registration matrix, a registration score corresponding to the optimal registration sequence and a registration vertex pair group of the optimal registration sequence and the second interior angle sequence; wherein the registration vertex pair group comprises a plurality of registration interior angle pairs and a corresponding plurality of interior angle correspondences; screening a first optimal matching sequence with the highest degree of matching from the optimal matching sequences according to the number of second internal angles included in each optimal matching sequence and the corresponding matching score; and taking the first optimal matching sequence, the registration vertex pair group and the corresponding first reference image spot as a first registration result.
The invention also provides an automatic registration system of the global comprehensive improvement image spots, which is characterized by comprising a user input module, an automatic registration module and a result generation module, wherein the automatic registration module is respectively connected to the user input module and the result generation module; the automatic registration module is used for executing the automatic registration method of the global comprehensive remediated image spots.
Compared with the prior art, the embodiment of the invention has the following beneficial effects:
the invention provides an automatic registration method, device and system for global comprehensive treatment of spots, which comprises the steps of performing registration scoring and score backtracking on inner angles in a first inner angle sequence and a second inner angle sequence which are registered by applying a Smith-Waterman algorithm and a vacancy weight constant model, and calculating an affine parameter group between a first reference spot and a first to-be-registered spot after screening and obtaining a first registration result with the highest registration degree and the first reference spot corresponding to the first registration result, so as to finish registration.
Further, the automatic registration method, device and system for the global comprehensive improvement of the image spots further determine whether the first registration result passes the verification by carrying out affine matching on the first reference image spot and the image spot to be registered, so that the reliability of the registration result is improved.
Drawings
The invention will be further described with reference to the accompanying drawings, in which:
FIG. 1 illustrates a flow diagram of one embodiment of a method for automatic registration of globally integrated healed patches according to the present invention;
FIG. 2 illustrates a flow diagram of another embodiment of a method for automatic registration of globally integrated healed patches according to the present invention;
FIG. 3 is a block diagram of an automatic registration apparatus for global integrated healing of patches according to the present invention;
fig. 4 shows a block diagram of an automatic registration system for globally integrated rectified patches according to 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.
Detailed description of the preferred embodiment
The embodiment of the invention firstly provides an embodiment of an automatic registration method of a global comprehensive improvement image spot. Fig. 1 shows a flow chart of an embodiment of an automatic registration method of a global integrated healing patch according to the present invention. As shown in fig. 1, the method comprises the steps of:
and S1, acquiring a first internal angle sequence of each of the plurality of reference image spots and a second internal angle sequence of the first image spot to be registered.
To achieve an automatic registration of the image spots, first a first to-be-registered image spot to be registered and a plurality of reference image spots for registration should be determined. Wherein a first length m of the first sequence of interior angles is not equal to a second length n of the second sequence of interior angles. In fact, because the spot boundaries are closed, its angular sequence has periodicity. Thus, in one embodiment, the second sequence of interior angles is extended to a sequence of interior angles to be matched of length 2m-1, in order to not miss possible matching points.
And S2, calculating and screening to obtain a first registration result with the highest registration degree and a first reference pattern spot corresponding to the first registration result according to the plurality of first interior angle sequences, the second interior angle sequence, the Smith-Watermann algorithm and the vacancy weight constant model.
Specifically, according to a Smith-Waterman algorithm and a vacancy weight constant model, sequentially aiming at each first interior angle sequence, respectively calculating a matching score between each first interior angle in the first interior angle sequence and each second interior angle in the second interior angle sequence, and constructing a registration matrix according to the matching scores; respectively backtracking the registration matrixes to respectively obtain an optimal registration sequence corresponding to each registration matrix, a registration score corresponding to the optimal registration sequence and a registration vertex pair group of the optimal registration sequence and the second interior angle sequence; screening a first optimal matching sequence with the highest degree of matching from the optimal matching sequences according to the number of second internal angles included in each optimal matching sequence and the corresponding matching score; and taking the first optimal matching sequence, the registration vertex pair group and the corresponding first reference image spot as a first registration result. Wherein the registration vertex pair group includes a plurality of registration internal angle pairs and a corresponding plurality of internal angle correspondences, and the first registration result includes a registration vertex pair group.
In one embodiment, the calculation formula of the matching score is specifically:
in the formula, HijIs the ith first interior angle a in the sequence of first interior anglesiAnd the jth second interior angle c in the sequence of second interior anglesjMatching score of (a), S (a)i,cj) Is the vertex internal angle aiAnd the vertex internal angle cjIs similar to the score of, W1A gap penalty can be set as required; δ is a preset spatial distance. In one embodiment, the gap penalty may be set to 2.
In an embodiment, the process of backtracking the registration matrix specifically includes: backtracking from the item with the largest score in the registration matrix; then, writing an optimal registration sequence according to the backtracking path; and finally, taking the item with the largest score in the registration matrix as the matching score of the optimal registration sequence, and determining the registration vertex pair group of the optimal registration sequence and the second internal angle sequence.
In the process, the backtracking comprises the following specific steps: if ai-cjIf the | is less than or equal to delta, backtracking to the upper left cell; if ai-cj|>Delta, backtracking to the cell with the maximum median at the upper left corner, the upper edge and the left side (if the cell with the same maximum value exists, the priority is according to the order of the upper left corner, the upper edge and the left side); and ending the backtracking process until the element with the score of zero is backtracked.
In the above process, the specific steps of writing out the optimal registration sequence according to the backtracking path are as follows: if backtracking to the upper left corner cell, aiIs added to the optimal registration sequence a', c is addedjAdding the sequence number j of the index into the matching index sequence C'; if the upper edge cell is traced back, a isiAdd sequence number i to the optimal registration sequence a ', add _ to the matching index sequence C'; if backtracking to the left cell, add _ to the optimal registration sequence A', add cjSequence number j is added to the matching index sequence C'.
The establishing of the registration matrix according to the matching score specifically comprises: respectively constructing a zero matrix according to the length of each first interior angle sequence and the length of each second interior angle sequence; and respectively filling the matching scores of each first internal angle and each second internal angle into corresponding positions of the corresponding zero matrix by taking the serial number of the first internal angle as a column coordinate and the serial number of the second internal angle as a row coordinate aiming at each zero matrix and the corresponding first internal angle sequence, so as to respectively obtain the registration matrix corresponding to each first internal angle sequence. The plurality of zero matrixes correspond to the plurality of first interior angle sequences one by one.
And S3, calculating an affine parameter group between the first reference image spot and the first to-be-registered image spot according to the registration vertex pair group, thereby completing registration.
The invention provides an automatic registration method, device and system for global comprehensive treatment of spots, which comprises the steps of performing registration scoring and score backtracking on inner angles in a first inner angle sequence and a second inner angle sequence which are registered by applying a Smith-Waterman algorithm and a vacancy weight constant model, and calculating an affine parameter group between a first reference spot and a first to-be-registered spot after screening and obtaining a first registration result with the highest registration degree and the first reference spot corresponding to the first registration result, so as to finish registration.
Detailed description of the invention
Furthermore, the invention also provides another embodiment of the automatic registration method of the global comprehensive improvement image spots. Fig. 2 shows a flow chart of another embodiment of an automatic registration method of a global integrated healing patch according to the present invention. As shown in fig. 2, the method comprises the steps of:
a1, acquiring a reference layer and a layer to be registered, and overlapping the reference layer and the layer to be registered.
A2, selecting a first pattern spot to be registered from the pattern layer to be registered, and selecting a plurality of reference pattern spots from the reference pattern layer by taking the first pattern spot to be registered as the center according to a preset spatial distance.
A3, acquiring a first internal angle sequence of each of the plurality of reference image spots and a second internal angle sequence of the first image spot to be registered.
To achieve an automatic registration of the image spots, first a first to-be-registered image spot to be registered and a plurality of reference image spots for registration should be determined. Wherein a first length m of the first sequence of interior angles is not equal to a second length n of the second sequence of interior angles. In fact, because the spot boundaries are closed, its angular sequence has periodicity. Thus, in one embodiment, the second sequence of interior angles is extended to a sequence of interior angles to be matched of length 2m-1, in order to not miss possible matching points.
And A4, calculating and screening to obtain a first registration result with the highest registration degree and a first reference pattern spot corresponding to the first registration result according to the plurality of first interior angle sequences, the second interior angle sequence, the Smith-Watermann algorithm and the vacancy weight constant model.
Specifically, according to a Smith-Waterman algorithm and a vacancy weight constant model, sequentially aiming at each first interior angle sequence, respectively calculating a matching score between each first interior angle in the first interior angle sequence and each second interior angle in the second interior angle sequence, and constructing a registration matrix according to the matching scores; respectively backtracking the registration matrixes to respectively obtain an optimal registration sequence corresponding to each registration matrix, a registration score corresponding to the optimal registration sequence and a registration vertex pair group of the optimal registration sequence and the second interior angle sequence; screening a first optimal matching sequence with the highest degree of matching from the optimal matching sequences according to the number of second internal angles included in each optimal matching sequence and the corresponding matching score; and taking the first optimal matching sequence, the registration vertex pair group and the corresponding first reference image spot as a first registration result. Wherein the registration vertex pair group includes a plurality of registration internal angle pairs and a corresponding plurality of internal angle correspondences, and the first registration result includes a registration vertex pair group.
In one embodiment, the calculation formula of the matching score is specifically:
in the formula, HijIs the ith first interior angle a in the sequence of first interior anglesiAnd the jth second interior angle c in the sequence of second interior anglesjMatching score of (a), S (a)i,cj) Is the vertex internal angle aiAnd the vertex internal angle cjIs similar to the score of, W1A gap penalty can be set as required; δ is a preset spatial distance. In one embodiment, the gap penalty may be set to 2.
In an embodiment, the process of backtracking the registration matrix specifically includes: backtracking from the item with the largest score in the registration matrix; then, writing an optimal registration sequence according to the backtracking path; and finally, taking the item with the largest score in the registration matrix as the matching score of the optimal registration sequence, and determining the registration vertex pair group of the optimal registration sequence and the second internal angle sequence.
In the process, the backtracking comprises the following specific steps: if ai-cjIf the | is less than or equal to delta, backtracking to the upper left cell; if ai-cj|>Delta, backtracking to the cell with the maximum median at the upper left corner, the upper edge and the left side (if the cell with the same maximum value exists, the priority is according to the order of the upper left corner, the upper edge and the left side); and ending the backtracking process until the element with the score of zero is backtracked.
In the above process, the specific steps of writing out the optimal registration sequence according to the backtracking path are as follows: if backtracking to the upper left corner cell, aiIs added to the optimal registration sequence a', c is addedjAdding the sequence number j of the index into the matching index sequence C'; if the upper edge cell is traced back, a isiAdd sequence number i to the optimal registration sequence a ', add _ to the matching index sequence C'; if backtracking to the left cell, add _ to the optimal registration sequence A', add cjSequence number j is added to the matching index sequence C'.
The establishing of the registration matrix according to the matching score specifically comprises: respectively constructing a zero matrix according to the length of each first interior angle sequence and the length of each second interior angle sequence; and respectively filling the matching scores of each first internal angle and each second internal angle into corresponding positions of the corresponding zero matrix by taking the serial number of the first internal angle as a column coordinate and the serial number of the second internal angle as a row coordinate aiming at each zero matrix and the corresponding first internal angle sequence, so as to respectively obtain the registration matrix corresponding to each first internal angle sequence. The plurality of zero matrixes correspond to the plurality of first interior angle sequences one by one.
And A5, performing affine on the first reference image spot and the image spot to be registered according to the plurality of registration vertex pairs in the registration vertex pair group, and determining whether the first registration result passes verification according to an affine result.
A61: if not, stopping the automatic registration process and sending a registration failure prompt to the user.
In order to ensure the reliability of an automatic registration result, after a first reference image spot matched with a first image spot to be registered is determined, affine is carried out on the first reference image spot and the image spot to be registered according to a plurality of registration vertex pairs in the registration vertex pair group, and whether the first registration result passes verification is determined according to an affine result. When the registration fails, stopping the automatic registration process to avoid wasting computing resources, and sending a registration failure prompt to the user. In one embodiment, the registration failure prompt includes one or more of a voice, a beep, a textual message, a short video, and a logo.
And A62, if yes, calculating an affine parameter group between the first reference image spot and the first to-be-registered image spot according to the registration vertex pair group, thereby finishing registration.
The invention provides an automatic registration method, device and system of global comprehensive treatment image spots, which comprises the steps of performing registration scoring and score backtracking on inner angles in a first inner angle sequence and a second inner angle sequence which are registered by applying a Smith-Wattman algorithm and a vacancy weight constant model, and calculating an affine parameter group between a first reference image spot and a first image spot to be registered after screening and obtaining a first registration result with the highest registration degree and the first reference image spot corresponding to the first registration result, thereby completing registration; further, the automatic registration method, device and system for the global comprehensive improvement of the image spots further determine whether the first registration result passes the verification by carrying out affine matching on the first reference image spot and the image spot to be registered, so that the reliability of the registration result is improved.
Detailed description of the preferred embodiment
In addition to the above method, an embodiment of the present invention further provides an automatic registration apparatus for global integrated adjustment of an image spot. Fig. 3 is a block diagram illustrating an automatic registration apparatus for global integrated healing of patches according to the present invention.
As shown in fig. 3, the automatic registration apparatus includes a spot acquisition unit 11, a registration filtering unit 12, and a parameter calculation unit 13.
The pattern spot obtaining unit 11 is configured to obtain a first interior angle sequence of each of the plurality of reference pattern spots and a second interior angle sequence of the pattern spot to be aligned. Wherein a first length m of the first sequence of interior angles is not equal to a second length n of the second sequence of interior angles.
The registration screening unit 12 is configured to calculate and screen to obtain a first registration result with the highest registration degree and a first reference pattern corresponding to the first registration result according to the plurality of first interior angle sequences, the second interior angle sequence, the smith-waterman algorithm, and the vacancy weight invariant model. The first registration result comprises the number of matched vertexes and the registration vertex pair group.
The parameter calculating unit 13 is configured to calculate an affine parameter set between the first reference pattern spot and the pattern spot to be registered according to the registration vertex pair group, so as to complete registration.
When the automatic registration of the pattern spots to be registered is required, firstly, a pattern spot acquisition unit 11 is used for acquiring a first internal angle sequence of each of a plurality of reference pattern spots and a second internal angle sequence of the pattern spots to be registered; subsequently, the registration screening unit 12 calculates and screens to obtain a first registration result with the highest registration degree and a first reference pattern corresponding to the first registration result according to the plurality of first interior angle sequences, the second interior angle sequence, the smith-waterman algorithm and the vacancy weight constant model; finally, the parameter calculating unit 13 calculates an affine parameter set between the first reference pattern spot and the pattern spot to be registered according to the registration vertex pair group, thereby completing registration.
In one embodiment, the automatic registration apparatus further comprises a registration verification unit 14, the registration verification unit 14 being configured to: after the registration screening unit 12 calculates and screens a first registration result with the highest registration degree and a first reference pattern corresponding to the first registration result according to the plurality of first interior angle sequences, the second interior angle sequence, the smith-waterman algorithm and the vacancy weight constant model to obtain the first registration result with the highest registration degree, performing affine on the first reference pattern and the pattern to be registered according to a plurality of registration vertex pairs in the registration vertex pair group, and determining whether the first registration result passes verification according to the affine result; if not, stopping the automatic registration process and sending a registration failure prompt to the user.
In an embodiment, the automatic registration apparatus further comprises an object selection unit 15 for: before acquiring a first interior angle sequence of each of a plurality of reference pattern spots and a second interior angle sequence of a pattern spot to be registered through a pattern spot acquisition unit 11, acquiring a reference pattern layer and a pattern layer to be registered, and overlapping the reference pattern layer and the pattern layer to be registered; and selecting a first pattern spot to be registered from the layer to be registered, and selecting a plurality of reference pattern spots from the reference layer by taking the first pattern spot to be registered as a center according to a preset spatial distance.
The invention provides an automatic registration device of a global comprehensive treatment image spot, which is characterized in that a Smith-Waterman algorithm and a vacancy weight constant model are used for carrying out registration scoring and fraction backtracking on inner angles in a first inner angle sequence and a second inner angle sequence which are registered, after a first registration result with the highest registration degree and a first reference image spot corresponding to the first registration result are obtained by screening, an affine parameter group between the first reference image spot and a first image spot to be registered is calculated, so that registration is completed, and the automation and the high efficiency of the registration process of the global comprehensive treatment image spot are improved; furthermore, the automatic registration device for the global comprehensive improvement of the image spots further determines whether the first registration result passes the verification or not by carrying out affine matching on the first reference image spot and the image spot to be registered, so that the reliability of the registration result is improved.
Detailed description of the invention
In addition to the above method and apparatus, an embodiment of an automatic registration system for global integrated adjustment of patches is provided in the embodiments of the present invention. Figure 4 shows a block diagram of one embodiment of an automatic registration system for globally integrated rectified patches according to the present invention. As shown in fig. 4, the automatic registration system includes a user input module 2, an automatic registration module 1, and a result generation module 3, wherein the automatic registration module 1 is communicatively connected to the user input module 2 and the result generation module 3, respectively.
The automatic registration module 1 is configured to perform the automatic registration method for the global integrated healing patches as described above, so as to obtain a first reference patch and a corresponding affine parameter set.
The user input module 2 is used for inputting or selecting a reference layer and a layer to be registered in an existing database. In one embodiment, the user input module 2 includes a combination of one or more of a touchable/non-touchable display screen, an input keyboard, a virtual keyboard, and a touchscreen.
The result generation module 3 is used for sending prompt information of successful automatic registration to the user when the first registration result is judged to pass the verification according to the affine result between the first reference pattern spot and the pattern spot to be registered; and sending a registration failure prompt to the user when the first registration result is judged to be not verified according to the affine result. In one embodiment, the result generating module 3 is in communication connection with the user input module 2, so that when the affine result between the first reference pattern spot and the pattern spot to be registered is verified, the automatic registration result is sent to the user through the user input module 2; and when the affine result is not verified, sending a registration failure prompt to the user. In one embodiment, the automatic registration success hint information includes a registration success hint, a first reference blob, and a corresponding set of affine parameters.
When automatic registration is needed, a user inputs the information through the user input module 2 or selects a reference layer and a layer to be registered in the existing database; subsequently, the automatic registration module 1 executes the automatic registration method of the global comprehensive rectified patches as described above according to the reference layer and the layer to be registered, so as to obtain the first reference patch and the corresponding affine parameter set.
In one embodiment, the result generating module 3 sends an automatic registration success prompt message to the user when the automatic registration module 1 judges that the first registration result passes the verification according to the affine result between the first reference pattern spot and the pattern spot to be registered, so as to obtain the first reference pattern spot and the corresponding affine parameter set; and when the automatic registration module 1 judges that the first registration result is not verified, sending a registration failure prompt to the user.
The invention provides an automatic registration system of a global comprehensive treatment image spot, which is characterized in that a Smith-Waterman algorithm and a vacancy weight constant model are used for carrying out registration scoring and fraction backtracking on inner angles in a first inner angle sequence and a second inner angle sequence which are registered, after a first registration result with the highest registration degree and a first reference image spot corresponding to the first registration result are obtained by screening, an affine parameter group between the first reference image spot and a first image spot to be registered is calculated, so that registration is completed, and the automation and the high efficiency of the registration process of the global comprehensive treatment image spot are improved by the automatic registration system; furthermore, the automatic registration system for the global comprehensive improvement of the image spots further determines whether the first registration result passes the verification or not by carrying out affine matching on the first reference image spot and the image spot to be registered, so that the reliability of the registration result is improved.
The above-mentioned embodiments are provided to further explain the objects, technical solutions and advantages of the present invention in detail, and it should be understood that the above-mentioned embodiments are only examples of the present invention and are not intended to limit the scope of the present invention. It should be understood that any modifications, equivalents, improvements and the like, which come within the spirit and principle of the invention, may occur to those skilled in the art and are intended to be included within the scope of the invention.
Claims (8)
1. An automatic registration method for a global integrated regulation image spot, the automatic registration method comprising:
acquiring a first internal angle sequence of each of the plurality of reference image spots and a second internal angle sequence of the first image spot to be registered; wherein a first length m of the first sequence of interior angles is not equal to a second length n of the second sequence of interior angles;
calculating and screening to obtain a first registration result with the highest registration degree and a first reference pattern spot corresponding to the first registration result according to the plurality of first interior angle sequences, the second interior angle sequence, the Smith-Wattman algorithm and a vacancy weight constant model; the first registration result comprises a registration vertex pair group; the calculating and screening to obtain a first registration result with the highest registration degree and a first reference pattern spot corresponding to the first registration result according to the plurality of first interior angle sequences, the second interior angle sequence, the smith-waterman algorithm and the vacancy weight constant model specifically comprises: according to a Smith-Waterman algorithm and a vacancy weight constant model, sequentially aiming at each first interior angle sequence, respectively calculating the matching score between each first interior angle in the first interior angle sequence and each second interior angle in the second interior angle sequence, and constructing a registration matrix according to the matching scores; respectively backtracking the registration matrixes to respectively obtain an optimal registration sequence corresponding to each registration matrix, a registration score corresponding to the optimal registration sequence and a registration vertex pair group of the optimal registration sequence and the second interior angle sequence; wherein the registration vertex pair group comprises a plurality of registration interior angle pairs and a corresponding plurality of interior angle correspondences; screening a first optimal matching sequence with the highest degree of matching from the optimal matching sequences according to the number of second internal angles included in each optimal matching sequence and the corresponding matching score; taking the first optimal matching sequence, the registration vertex pair group and the corresponding first reference image spot as a first registration result;
and according to the registration vertex pair group, calculating an affine parameter group between the first reference image spot and the first to-be-registered image spot, thereby completing registration.
2. The automatic registration method for global integrated healing patches as claimed in claim 1, wherein after the step of performing a screening to obtain the first registration result with the highest degree of registration and the first reference patch corresponding to the first registration result, the automatic registration method further comprises:
performing affine on the first reference image spot and the image spot to be registered according to a plurality of registration vertex pairs in the registration vertex pair group, and determining whether the first registration result passes verification according to an affine result;
if not, stopping the automatic registration process and sending a registration failure prompt to the user.
3. The method of automatic registration of globally integrated and healed patches according to claim 2, wherein before obtaining the first sequence of internal angles of each of the plurality of reference patches and the second sequence of internal angles of the patch to be registered, the method further comprises:
acquiring a reference layer and a layer to be registered, and overlapping the reference layer and the layer to be registered;
and selecting a first pattern spot to be registered from the layer to be registered, and selecting a plurality of reference pattern spots from the reference layer by taking the first pattern spot to be registered as a center according to a preset spatial distance.
4. The automatic registration method for global integrated healing patches according to claim 3, wherein the registration matrix is constructed according to the matching scores, specifically:
respectively constructing a zero matrix according to the length of each first interior angle sequence and the length of each second interior angle sequence; the plurality of zero matrixes correspond to the plurality of first interior angle sequences one by one;
and respectively filling the matching scores of each first internal angle and each second internal angle into corresponding positions of the corresponding zero matrix by taking the serial number of the first internal angle as a column coordinate and the serial number of the second internal angle as a row coordinate aiming at each zero matrix and the corresponding first internal angle sequence, so as to respectively obtain the registration matrix corresponding to each first internal angle sequence.
5. An automatic registration device of global comprehensive improvement image spots is characterized by comprising an image spot acquisition unit, a registration screening unit and a parameter calculation unit,
the pattern spot acquisition unit is used for acquiring a first interior angle sequence of each of the plurality of reference pattern spots and a second interior angle sequence of the pattern spot to be registered; wherein a first length m of the first sequence of interior angles is not equal to a second length n of the second sequence of interior angles;
the registration screening unit is used for calculating and screening to obtain a first registration result with the highest registration degree and a first reference pattern spot corresponding to the first registration result according to the plurality of first interior angle sequences, the second interior angle sequence, the Smith-Wattman algorithm and a vacancy weight constant model; the first registration result comprises the number of matched vertexes and a registration vertex pair group; according to the plurality of first interior angle sequences, the second interior angle sequence, the Smith-Watermann algorithm and the vacancy weight constant model, calculating and screening to obtain a first registration result with the highest registration degree and a first reference pattern spot corresponding to the first registration result, specifically: according to a Smith-Waterman algorithm and a vacancy weight constant model, sequentially aiming at each first interior angle sequence, respectively calculating the matching score between each first interior angle in the first interior angle sequence and each second interior angle in the second interior angle sequence, and constructing a registration matrix according to the matching scores; respectively backtracking the registration matrixes to respectively obtain an optimal registration sequence corresponding to each registration matrix, a registration score corresponding to the optimal registration sequence and a registration vertex pair group of the optimal registration sequence and the second interior angle sequence; wherein the registration vertex pair group comprises a plurality of registration interior angle pairs and a corresponding plurality of interior angle correspondences; screening a first optimal matching sequence with the highest degree of matching from the optimal matching sequences according to the number of second internal angles included in each optimal matching sequence and the corresponding matching score; taking the first optimal matching sequence, the registration vertex pair group and the corresponding first reference image spot as a first registration result;
and the parameter calculating unit is used for calculating an affine parameter group between the first reference pattern spot and the pattern spot to be registered according to the registration vertex pair group, so that registration is completed.
6. The apparatus of automatic registration of global integrated healed patches according to claim 5, further comprising a registration verification unit for:
performing affine on the first reference image spot and the image spot to be registered according to a plurality of registration vertex pairs in the registration vertex pair group, and determining whether the first registration result passes verification according to an affine result;
if not, stopping the automatic registration process and sending a registration failure prompt to the user.
7. The apparatus of automatic registration of global integrated healed patches according to claim 6, further comprising an object selection unit for:
acquiring a reference layer and a layer to be registered, and overlapping the reference layer and the layer to be registered;
and selecting a first pattern spot to be registered from the layer to be registered, and selecting a plurality of reference pattern spots from the reference layer by taking the first pattern spot to be registered as a center according to a preset spatial distance.
8. An automatic registration system of global integrated healed patches, characterized in that the automatic registration system comprises a user input module, an automatic registration module and a result generation module, the automatic registration module is respectively connected to the user input module and the result generation module; the automatic registration module is used for executing the automatic registration method of the globally integrated and healed image spots according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110838660.7A CN113610700B (en) | 2021-07-23 | 2021-07-23 | Automatic registration method, device and system for global comprehensive improvement of image spots |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110838660.7A CN113610700B (en) | 2021-07-23 | 2021-07-23 | Automatic registration method, device and system for global comprehensive improvement of image spots |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113610700A CN113610700A (en) | 2021-11-05 |
| CN113610700B true CN113610700B (en) | 2022-03-22 |
Family
ID=78305326
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110838660.7A Active CN113610700B (en) | 2021-07-23 | 2021-07-23 | Automatic registration method, device and system for global comprehensive improvement of image spots |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113610700B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109934866A (en) * | 2019-03-07 | 2019-06-25 | 武大吉奥信息技术有限公司 | A kind of extracting method, copy method and the device of figure spot partition mode |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SE9903988D0 (en) * | 1999-11-03 | 1999-11-03 | Amersham Pharm Biotech Ab | Method of analysis |
| CN101655982B (en) * | 2009-09-04 | 2011-06-29 | 上海交通大学 | Image registration method based on improved Harris angular point |
| CN111127634B (en) * | 2019-12-20 | 2023-04-28 | 山东省国土测绘院 | Method, device, equipment and storage medium for superimposing pattern spots of land utilization data |
-
2021
- 2021-07-23 CN CN202110838660.7A patent/CN113610700B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109934866A (en) * | 2019-03-07 | 2019-06-25 | 武大吉奥信息技术有限公司 | A kind of extracting method, copy method and the device of figure spot partition mode |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113610700A (en) | 2021-11-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109614935B (en) | Vehicle damage assessment method and device, storage medium and electronic equipment | |
| CN110907947B (en) | Real-time loop detection method in mobile robot SLAM problem | |
| CN106338989B (en) | A kind of field robot binocular vision navigation methods and systems | |
| CN111046717A (en) | Fundus image macular center positioning method and device, electronic equipment and storage medium | |
| CN102721364B (en) | A kind of localization method of workpiece and device thereof | |
| CN108827249A (en) | A kind of map constructing method and device | |
| CN109725327A (en) | A kind of method and system of multimachine building map | |
| CN105809107A (en) | Single-sample face identification method and system based on face feature point | |
| CN106897677B (en) | Vehicle feature classification retrieval system and method | |
| CN111652057A (en) | Map construction method and device, computer equipment and storage medium | |
| CN109117851A (en) | A kind of video image matching process based on lattice statistical constraint | |
| CN109214314A (en) | A kind of lane line merges matching algorithm automatically | |
| CN109993108B (en) | Gesture error correction method, system and device under a kind of augmented reality environment | |
| Yu et al. | Accurate and robust visual localization system in large-scale appearance-changing environments | |
| CN113610700B (en) | Automatic registration method, device and system for global comprehensive improvement of image spots | |
| CN117273813B (en) | Project intelligent site selection method considering environment control | |
| CN109934866A (en) | A kind of extracting method, copy method and the device of figure spot partition mode | |
| CN109492522A (en) | Specific objective detection model training program, equipment and computer readable storage medium | |
| CN117152844A (en) | High-integrity worker construction attitude detection method and system based on computer vision | |
| CN112052572A (en) | Digital twin industrial simulation system based on WLAN (Wireless local area network) position perception | |
| CN110415171B (en) | Image processing method, image processing device, storage medium and electronic equipment | |
| CN113673288A (en) | Idle parking space detection method and device, computer equipment and storage medium | |
| CN112308842B (en) | Positioning core automatic extraction method based on printed matter image | |
| CN116798087A (en) | Employee on-duty state detection method and system | |
| CN115761516B (en) | Method, server and storage medium for analyzing landing zone of air emergency delivery |
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 |