Ship body section precision intelligent detection method based on photogrammetry technology

Abstract

The invention discloses an intelligent hull section precision detection method based on a photogrammetry technology, and belongs to the technical field of ship measurement. According to the method, an unmanned aerial vehicle is used for carrying out data collection on to-be-measured ship body segments, computer software is used for carrying out modeling on the collected data, and measurement and precision analysis work are carried out on a model. The method comprises the following steps of S100, environment data collection and S200, field data collection. S300, reconstructing a ship model; and S400, carrying out automatic analysis based on the ship model. According to the method, data acquisition is carried out on the to-be-measured ship body sections through the unmanned aerial vehicle, compared with traditional manual acquisition through a total station, the unmanned aerial vehicle and close photogrammetry are combined for the first time to be used in the ship measurement field, millimeter-level precision can be achieved, unmanned and intelligent effects are achieved, and whether actually-produced sections meet design requirements or not can be efficiently and accurately judged.

Description

Technical Field

[0001] The present invention belongs to the technical field of ship measurement, and specifically relates to an intelligent detection method for the accuracy of hull segments based on photogrammetry technology. Background Art

[0002] Currently, the main method used in shipbuilding is the sectional processing method. This method can measure the ship using a high-precision total station to ensure the accuracy of each segment during the welding and assembly process, thereby ensuring the accuracy of the overall installation and construction of the ship. At present, the hull segment accuracy detection method mainly uses ground measurement equipment such as total stations and laser scanners to collect data. By the way of dotting with the total station, the coordinate points to be spliced are collected, and then the accuracy of the currently manufactured ship segment is obtained by comparing the actually measured coordinates with the data coordinates designed for the ship. There are the following disadvantages in using such a method for hull segment accuracy measurement:

[0003] 1. It is necessary to install equipment manually and place a target on the object to be measured manually. However, since the height of general hull segments is relatively high, when placing the target, construction workers need to work at high altitudes and are prone to falling from high altitudes, and the safety factor of production operations is relatively low.

[0004] 2. The measurement operation has a high degree of manual participation, and the measurement accuracy is affected by the experience and technical level of the operators, and the accuracy of the measurement results is unstable.

[0005] 3. It requires a large amount of manpower;

[0006] 4. It is impossible to obtain the surface information and three-dimensional information of the ship segment, and only single-point information can be obtained. Cracks, paint peeling, and rust cannot be detected;

[0007] 5. It is impossible to achieve automated and unmanned data collection and analysis. Summary of the Invention

[0008] 1. Technical Problems to be Solved by the Invention

[0009] The purpose of the present invention is to solve the problem that it is difficult to achieve intelligence and automation in the existing ship segment measurement.

[0010] 2. Technical Solutions

[0011] To achieve the above purpose, the technical solution provided by the present invention is as follows:

[0012] An intelligent detection method for the accuracy of hull segments based on photogrammetry technology of the present invention. The method is to use an unmanned aerial vehicle to collect data of the hull segment to be measured, use computer software to model the collected data, and perform measurement and accuracy analysis work on the model. The method includes the following steps:

[0013] S100. Collect environmental data,

[0014] S200. Collect field data;

[0015] S300. Reconstruct the ship model;

[0016] S400. Automatically analyze based on the ship model.

[0017] Preferably, the step S100 is specifically to collect the environmental data of the area to be measured through historical accurate terrain data, three-dimensional laser scanning point cloud data or drones, and the environmental data is terrain data and building data.

[0018] Preferably, the step S100 further includes when collecting through drones, pre-flying the area to be measured in a grid pattern by drones in advance to obtain multi-angle aerial images; the requirements for setting the flight path of the drones are

[0019] a. Repeatedly shoot the measurement area with reciprocating flight paths according to the requirement of not less than 70% overlap of aerial photos (more than 75% in the heading direction and more than 70% in the side direction);

[0020] b. Plan at least 4 flight paths or 6 exposure point positions outside the effective area;

[0021] c. The flight path curvature is not greater than 3%.

[0022] Preferably, the collection of field data in the step S200 is specifically to collect the overall image data of the ship to be measured, and the collection methods of the collection of field data include drone close-range photography data collection, supplementary shooting of key areas, and field control point positioning collection.

[0023] Preferably, the drone close-range photography data collection specifically uses drones for close-range photography, and specifically includes the following steps:

[0024] a. Divide the area to be measured;

[0025] b. Design an ideal flight path;

[0026] c. Import the ideal flight path into the flight control of the drone to take pictures.

[0027] Preferably, the step a is specifically to realize the optimal acquisition of images guided by target features and the maximum expression of target information based on the intelligent planning of close-range photography three-dimensional flight tracks and attitudes adaptive to terrain and the optimized expression of the geometric structure of complex-shaped targets.

[0028] Preferably, the step b is specifically

[0029] If it is described by two sets of parameters, namely the midpoints (v1, v2) of the two short sides of the target outer contour rectangle and the width w of the rectangle, then the long side direction of the rectangle (i.e., the direction of the vector <v1, v2>) is the main direction of the shipping lane.

[0030] Preferably, the supplementary shooting of the key area is specifically to supplement the shooting of the ship area that is difficult for the UAV to directly and clearly shoot through manual control of the UAV. During the supplementary shooting process, auxiliary lines are set to shoot the object clearly from various angles by moving the shooting lens. The overlap rate between adjacent photos is controlled between 60% and 80%, and at least 3 adjacent photos capture the same point of the object being photographed.

[0031] Preferably, the automatic analysis based on the ship model in step S400 is specifically to fit the ship model generated in step S300 with the design file based on cad dxf, and generate a report on the comparison data.

[0032] Preferably, the fitting process is specifically to place the ship model and the design file in the same coordinate system and select several reference points for association, and obtain a visual report on whether the actual production segments meet the design requirements through comparison.

[0033] 3. Beneficial effects

[0034] Adopting the technical solution provided by the present invention, compared with the prior art, it has the following beneficial effects:

[0035] An intelligent detection method for the accuracy of hull segments based on photogrammetry technology of the present invention, the method is to use a UAV to collect data of the hull segments to be measured, use computer software to model the collected data, and perform measurement and accuracy analysis on the model. The method includes the following steps: S100, collecting environmental data; S200, collecting field data; S300, reconstructing the ship model; S400, automatically analyzing based on the ship model. By using a UAV to collect data of the hull segments to be measured, compared with the traditional manual collection using a total station, for the first time, the UAV and close-range photogrammetry are combined and used in the field of ship measurement, which can achieve millimeter-level accuracy, and is unmanned, intelligent, and can efficiently and accurately judge whether the actual production segments meet the design requirements. Description of the drawings

[0036] Figure 1 It is the overall flowchart of an intelligent detection method for the accuracy of hull segments based on photogrammetry technology of the present invention;

[0037] Figure 2 It is the schematic diagram of the technical process of Embodiment 1;

[0038] Figure 3 It is the technical route of UAV close-range photogrammetry data collection in Embodiment 1;

[0039] Figure 4 Schematic flow chart of model reconstruction for Embodiment 1;

[0040] Figure 5 Schematic diagram of the target outer contour approaching the flight path planning for Embodiment 1;

[0041] Figure 6 Flow chart of three-dimensional refined modeling for Embodiment 1;

[0042] Figure 7 Schematic diagram of UAV data collection for Embodiment 1;

[0043] Figure 8 Schematic diagram of the segmented approach route planning of the UAV for Embodiment 1 Figure 1 ;

[0044] Figure 9 Schematic diagram of the segmented approach route planning for Embodiment 1 Figure 2 。 Detailed implementation manners

[0045] In order to enable those skilled in the art to better understand the solution of this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of this application.

[0046] It should be noted that the terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so as to describe the embodiments of this application here. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0047] In this application, the orientation or positional relationships indicated by the terms "upper", "lower", "left", "right", "front", "rear", "top", "bottom", "inner", "outer", "middle", "vertical", "horizontal", "lateral", "longitudinal", etc. are based on the orientation or positional relationships shown in the drawings. These terms are mainly used to better describe this application and its embodiments, and are not used to limit that the indicated devices, elements or components must have a specific orientation, or be constructed and operated in a specific orientation.

[0048] Moreover, in addition to being able to represent orientation or positional relationships, some of the above terms may also be used to represent other meanings. For example, the term "upper" may also be used to represent a certain attachment relationship or connection relationship in some cases. For those of ordinary skill in the art, the specific meanings of these terms in this application can be understood according to specific circumstances.

[0049] In addition, the terms "install", "set", "provided with", "connect", "connected", "socketed" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral structure; it can be a mechanical connection or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, or there is internal communication between two devices, elements or components. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0050] It should be noted that, without conflict, the embodiments in this application and the features in the embodiments can be combined with each other. The following will detail this application with reference to the drawings and in combination with the embodiments.

[0051] Embodiment 1

[0052] Referring to the attached Figure 1-9 , a method for intelligent detection of the accuracy of a ship hull section based on photogrammetry technology in this embodiment. The method is to use a drone to collect data of the ship hull section to be measured, use computer software to model the collected data, and perform measurement and accuracy analysis work on the model. The method includes the following steps:

[0053] S100. Collect environmental data,

[0054] S200. Collect field data;

[0055] S300. Reconstruct the ship model;

[0056] S400. Automatically analyze based on the ship model.

[0057] In this embodiment, a drone is used to collect data of the ship hull section to be measured. Compared with the traditional manual collection using a total station, it is safer and more efficient, realizing unmanned and intelligent collection.

[0058] Specifically, step S100 is to collect environmental data of the area to be measured through historical accurate terrain data, 3D laser scanning point cloud data, or drones. The environmental data includes terrain data and building data.

[0059] This embodiment specifically is

[0060] Based on project requirements, the actual natural geographical environment of the measurement area should be surveyed. Before the survey, materials such as traffic maps, topographic maps, and human environments of the measurement area need to be collected. For the key areas of the shipyard segments, conduct on-site surveys under the leadership of internal staff to understand the key points of information collection and matters that need attention.

[0061] Determine the measurement area according to the collected materials and project requirements. Generally, expand the collection range to a certain extent according to the terrain and building conditions.

[0062] According to the field reconnaissance situation, estimate the workload, construction period, and the number of personnel for data collection and processing, and determine the data collection equipment to ensure the orderly connection of each process.

[0063] Step S100 also includes when collecting through drones, pre-flying the area to be measured in a grid pattern by drones to obtain multi-angle aerial images. The requirements for the flight path settings of the drones are

[0064] a. Repeatedly shoot the measurement area along the reciprocating flight path according to the requirement that the overlap degree of the aerial photos is not less than 70% (more than 75% in the flight direction and more than 70% in the side direction);

[0065] b. Plan at least 4 flight paths or 6 exposure point positions outside the effective area;

[0066] c. The curvature of the flight path is not greater than 3%.

[0067] After the acquisition of the external field data of the aerial photos is completed, first, conduct a quality inspection on the acquired images, and re-fly the unqualified areas until the quality of the acquired images meets the requirements; when the photo data meets the requirements, all the photo data can be imported into the modeling software.

[0068] Export the original images and POS data, set the geodetic reference system to the WGS-84 coordinate system, and import the camera calibration parameters. Through aerial triangulation encryption processing, recalculate the image orientation and attitude at the moment of photography.

[0069] Automatically output 3D point cloud LAS data using the results of aerotriangulation encryption. There are many photogrammetry systems that can perform aerotriangulation encryption to produce LAS point cloud result data, such as DPGrid of the School of Remote Sensing and Information Engineering of Wuhan University, ContextCapture software of Bentley, Photoscan software developed by Agisoft of Russia, and DJI Smart Mapping produced by DJI in China. All these software can perform mature processing. Since the final output result requires an ultra-fine 3D model, considering automation and production efficiency, DJI Smart Mapping software is recommended.

[0070] The collection of field data in step S200 specifically refers to collecting the overall image data of the ship to be measured. The collection methods of the field data collection method include UAV close-range photogrammetry data collection, supplementary shooting of key areas, and field control point positioning collection.

[0071] The UAV close-range photogrammetry data collection specifically uses a UAV for close-range photography, which specifically includes the following steps:

[0072] a. Divide the area to be measured;

[0073] b. Design an ideal flight path;

[0074] c. Import the ideal flight path into the flight control of the UAV to take pictures.

[0075] Step a specifically refers to the intelligent planning of 3D flight tracks and attitudes for close-range photography based on terrain adaptability, and the optimization of the geometric structure expression of complex-shaped objects, so as to achieve the optimal acquisition of images guided by target features and the maximum expression of target information.

[0076] Step b specifically refers to

[0077] Describe it with two sets of parameters: the midpoints (v1, v2) of the two short sides of the target outer contour rectangle and the width w of the rectangle. Then the long side direction of the rectangle (i.e., the direction of the vector <v1, v2>) is the main direction of the flight path.

[0078] The intelligent planning of 3D flight tracks and attitudes for close-range photography based on terrain adaptability, and the optimization of the geometric structure expression of complex-shaped objects, so as to achieve the optimal acquisition of images guided by target features and the maximum expression of target information. Through the processing strategy of "simplifying the complex", first divide the complex target, perform optimal design and processing on the segmented "surface" small targets, and combine the optimal solutions, sub-optimal solutions, etc. of multiple small targets with different weights to finally achieve the optimal solution of the overall target, that is, the best photography position, the best photography angle, the least number of photography times, and at the same time, the target can be expressed in the most detailed way.

[0079] To reduce the time the drone spends on route switching during shooting, the long side of the target's outer contour should be used as the main route direction, so that the number of routes required to cover the top surface of the target will be less. As Figure 5 As shown in (a), it is described by two parameters: the midpoints (v1, v2) of the two short sides of the rectangular target outer contour and the width w of the rectangle. Then, the long side direction of the rectangle (i.e., the direction of the vector <v1, v2>) is the main direction of the route.

[0080] When the shooting plane changes from the vertical plane to the horizontal plane, the distance between adjacent exposure points in the same flight track is still Δs, but the rotation angle φ of the drone's camera lens changes from 0° to -90°, and the distance Δh between adjacent exposure points in the vertical direction also correspondingly becomes the distance Δw between adjacent routes. Then, the number of routes required to shoot the top surface of the building in a vertically downward manner is:

[0081]

[0082] Among them, represents rounding up a upward.

[0083] The endpoint coordinates of these vertically downward shooting routes can be calculated according to the following formula:

[0084]

[0085] Among them, K represents the unit vector in the direction after the vector <v1, v2> is rotated 90° clockwise.

[0086] As Figure 5 As shown in (b), to ensure that the images in the inclined route can simultaneously capture the top surface and the side surface, the main optical axis of the camera needs to point to the edge of the top, that is, the two long sides of the rectangle. If the rotation angle of the camera lens in the inclined route is φ, then the endpoint coordinates of the two inclined routes are respectively:

[0087]

[0088] Obviously, the distance between adjacent exposure points in the inclined route is also Δs. Finally, along the direction of the vector K, sort the N1 + 2 routes calculated according to the distance from the straight line where the vector <v1, v2> is located, and combine them in a snake-shaped route to obtain the flight track planning result of the top surface. For other complex targets, they can be dissected into a combination of multiple rectangles, and the close flight track planning can be carried out respectively according to the method described above to obtain an ideal flight path.

[0089] Importing the ideal flight path into the flight control of the UAV can obtain comprehensive and detailed images of the target area. Combining with the RTK system carried by the aircraft, multi-scale data collection of the target area information can be realized. Integrating the three-dimensional real scene modeling technology can obtain more texture and structure information of the object, making the structure and texture of the object more detailed and the modeling effect more real and accurate.

[0090] The supplementary shooting of the key area specifically refers to manually controlling the UAV to supplement the shooting of the ship area that is difficult for the UAV to directly and clearly shoot. During the supplementary shooting process, auxiliary lines are set to clearly shoot each angle of the object by moving the shooting lens. The overlap rate between adjacent photos is controlled between 60% - 80%, and at least 3 adjacent photos are taken of the same point of the object being photographed. For complex objects, it is recommended that first, keep the lens horizontal with the center of the object and take a circle around the object, then take another circle 30 degrees above and below the horizontal respectively, and finally, take some additional photos of the top of the object and some supplementary photos of the local details of the complex object. In each photo, it is necessary to ensure that the object being photographed is entirely within the frame, and the object being photographed should occupy most of the frame, about 80% of the frame area, and a certain background needs to be reserved for subsequent processing.

[0091] Layout of control points is the most important work in the fieldwork of aerial surveying. The quality of photo control points will directly affect the absolute accuracy of the entire working area. In the embodiment, the UAVs used all have RTK systems with relatively high relative accuracy, so the number of control points can be appropriately reduced. When selecting control points, the requirements for point layout should be fully considered, and ground points and target points with good visibility for topographic surveying and can be clearly identified should be selected; control points need to be laid out around the entire survey area.

[0092] The model reconstruction in step S300 includes the following steps:

[0093] S310. Data inspection;

[0094] S320. Data preprocessing;

[0095] S330. Model reconstruction;

[0096] S400. Automatic analysis based on the ship model.

[0097] The data inspection in step S310 is mainly for the acquired frontal images, intersection images, and handheld'supplementary shooting' images. The main inspection items are as follows:

[0098] Whether there are relative and absolute holes in the close-range photography. If so, make up the photography in time, and both ends of the supplementary photography route should exceed two baselines outside the hole;

[0099] Check the image quality. The image should be clear, with moderate contrast, saturated color, vivid colors, and consistent color tone. Ensure that the color tone of the same ground objects is basically the same. It should have rich levels, be able to distinguish small ground object images suitable for the ground resolution, and be able to establish a clear stereo model. There should be no defects such as clouds, cloud shadows, smoke, large-area reflections, and stains on the image. Although there are a small number of defects, as long as they do not affect the connection of the stereo model and the establishment of the 3D model, it can be used for 3D model production.

[0100] Ensure that each image data corresponds to a POS positioning information data.

[0101] The operator manually visually checks whether the aerial flight data has the following situations: the photos are foggy, the lighting is uneven, there is noise, weak information, and color difference in the flight strip, etc. Usually, if the above situations are found in the images after inspection, it is necessary to re-fly and collect data under better weather conditions.

[0102] The data preprocessing in step S320 is specifically to adjust the color, brightness, and contrast of the original images and perform color homogenization before model reconstruction. The color homogenization should reduce the tonal differences between images, make the tone uniform, with moderate contrast and distinct levels, keep the colors of ground objects unchanged, and there should be no traces of color homogenization.

[0103] The model reconstruction is based on the data collected in the field, including ortho-view image data, cross-view image data, and local key area image data. It has the characteristics of large photographic tilt angle, serious image deformation; large resolution variation, unable to unify the scale; many overlaps, requiring multi-view processing, etc. Its data is different from that of conventional digital aerial photogrammetry. The modeling software used in this embodiment is DJI Terra, which currently meets the processing of multi-view aerial photography images. This software has a high degree of automation, few operation steps, and fast single-machine processing speed, and can well complete the aerial triangulation calculation. All photos will successfully participate in the calculation and quickly build the real-scene 3D model data.

[0104] Perform dense matching of multi-view image feature points on the camera parameters, image data, and POS data, and perform free network multi-view image joint constrained adjustment calculation for the regional network based on this to establish a stereo model that can be moderately freely deformed in the spatial scale and complete relative orientation; transfer and pierce the measured photo control point results in the fieldwork in the indoor environment, and use these points to perform constrained adjustment calculation on the existing regional network model to incorporate the regional network into the precise geodetic coordinate system and complete absolute orientation; directly submit the results data after aerial triangulation to generate 3D TIN grid construction, white body 3D model creation, automatic texture mapping, and 3D scene construction.

[0105] The automatic analysis based on the ship model in step S400 of this embodiment specifically refers to fitting the ship model generated in step S300 with the design file based on cad dxf, and generating a report on the comparison data.

[0106] Specifically, the fitting process specifically refers to placing the ship model and the design file in the same coordinate system and selecting several reference points for association, and obtaining a visual report on whether the actual production segments meet the design requirements through comparison.

[0107] The above embodiments only represent a certain implementation manner of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent for the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent for the present invention should be subject to the appended claims.

Claims

1. An intelligent detection method for hull segmentation accuracy based on photogrammetry technology, characterized by: The method is to use a drone to collect data on the hull segments to be measured, use computer software to model the collected data, and perform measurement and precision analysis on the model. The method includes the following steps: S100, collecting environmental data; S200, collect field data; S300, ship model reconstruction; S400, automatic analysis based on ship model.

2. According to claim 1, a method for intelligent detection of hull segmentation accuracy based on photogrammetry technology is characterized by: The step S100 specifically involves collecting environmental data of the area to be measured through historical precise terrain data, three-dimensional laser scanning point cloud data or drones, wherein the environmental data includes terrain data and building data.

3. The method for intelligent detection of hull segmentation accuracy based on photogrammetry technology according to claim 2 is characterized by: The step S100 also includes, when collecting data by drone, using the drone to fly in a tic-tac-toe pattern over the area to be measured in advance to obtain multi-angle aerial images; the route setting requirements of the drone are a. Repeatedly shoot the survey area on a reciprocating route according to the requirement that the aerial photographs have an overlap of no less than 70% (more than 75% in the heading direction and more than 70% in the lateral direction); b. Plan at least 4 routes or 6 exposure points outside the effective area; c. The curvature of the route shall not exceed 3%.

4. According to claim 2, a method for intelligent detection of hull segmentation accuracy based on photogrammetry technology is characterized by: The field data collection in step S200 specifically involves collecting the overall image data of the ship to be tested, and the field data collection methods include drone close-up photography data collection, key area supplementary shooting, and field control point positioning collection.

5. According to claim 4, a method for intelligent detection of hull segmentation accuracy based on photogrammetry technology is characterized in that: The drone close-up photography data collection specifically uses a drone to perform close-up photography, and specifically includes the following steps: a. Segment the area to be tested; b. Design the ideal flight path; c. Import the ideal flight path into the drone’s flight control system for filming.

6. The method for intelligent detection of hull segmentation accuracy based on photogrammetry technology according to claim 5 is characterized by: The step a is specifically based on terrain-adaptive close-up photography three-dimensional track and attitude intelligent planning, and optimized expression of the geometric structure of complex-shaped targets, so as to achieve optimal acquisition of target feature-oriented images and maximize expression of target information.

7. The method for intelligent detection of hull segmentation accuracy based on photogrammetry technology according to claim 5 is characterized by: The step b is specifically as follows: The target outer contour rectangle is described by two sets of parameters: the midpoints of the two short sides (v1, v2) and the width w of the rectangle. Then the direction of the long side of the rectangle (i.e., the vector<v1,v2> The direction of the route is the main direction of the route.

8. The method for intelligent detection of hull segmentation accuracy based on photogrammetry technology according to claim 4 is characterized by: The reshooting of key areas specifically involves reshooting the ship areas that are difficult for the drone to directly shoot clearly through manually controlled drones. During the reshooting process, auxiliary lines are set to clearly shoot the object from all angles by moving the shooting lens. The overlapping rate between adjacent photos is controlled between 60% and 80%, and at least 3 adjacent photos are taken of the same point of the photographed object.

9. The intelligent detection method of ship segmentation accuracy based on photogrammetry technology according to claim 1 is characterized by: The automatic analysis based on the ship model in step S400 is specifically to match the ship model generated in step S300 with the design file based on cad dxf, and generate a report based on the compared data.

10. The method for intelligent detection of ship segmentation accuracy based on photogrammetry technology according to claim 9, characterized in that: The integration process specifically includes placing the ship model and the design file in the same coordinate system and selecting several reference points for association, and comparing to obtain a visual report on whether the actually produced sections meet the design requirements.

Images