Device and method for digital image acquisition, digital image processing device and digital camera

Abstract

The invention relates to a device (40) for digital image acquisition and / or digital image output, in particular a digital camera device, with at least one image generation unit (10) which is at least configured to generate, in particular to record, at least two differently parameterized images (12, 14) of a scene (16, 18), in particular a completely real, completely artificial or partially artificial and partially real scene. It is proposed that the device (40) includes a digital image processing device (20) which is at least configured to combine the at least two generated, in particular recorded, differently parameterized images (12, 14) in such a way that a combined image (22) is generated by anisotropic and / or content-dependent digital image synthesis of the at least two images (12, 14), in particular modified at least in some areas compared to the original scene (16, 18), and preferably output to a user.

Description

State of the art

[0001] Devices for digital image capture, comprising at least one image generation unit configured to generate at least two differently parameterized images of a scene, such as multi-lens camera systems in smartphones, are known. Furthermore, simple zoom functionalities for such camera systems have also been known for a long time, particularly in the consumer electronics sector. There are now also many algorithms and applications that can manipulate image material and detect and uniquely assign specific components within it. Disclosure of the invention

[0002] The invention is based on a device for digital image recording and / or digital image output, preferably a digital camera device, with at least one image generation unit which is at least configured to generate, in particular to record, at least two differently parameterized images of a scene, in particular an at least partially overlapping, preferably identical scene.

[0003] It is proposed that the device comprise a digital image processing unit which is at least configured to combine the at least two generated, in particular recorded, differently parameterized images in such a way that, by means of anisotropic, in particular direction-dependent, and / or content-dependent digital image synthesis of the at least two images, a combined image is generated, in particular modified at least in regions compared to the original scene, and is preferably output to a user. This advantageously makes it possible to create a novel functionality for image recording devices, such as digital cameras or smartphones. Image recording and / or image output can advantageously be improved. The number of options for image processing and / or image optimization, in particular directly on image recording devices, can advantageously be increased.

[0004] Photographs can often include different components, such as a person in the foreground who is to be portrayed, and a landscape in the background. The physics of classic photography can be limiting, for example when a certain focal length has to be chosen for a shot which does not make it possible to optimally depict the subject, e.g. the person in the foreground, and the background, e.g. the landscape, in terms of relative size. It can also happen that photo subjects have different levels of interest in different directions, for example a mountain range in the distance, which should either be shot in its entire width, whereby the image is dominated by the sky or the nearby ground, or zoomed in strongly, which brings the mountain range into focus but only allows a narrow horizontal area to be captured.Such limitations can be advantageously overcome with the proposed device.

[0005] The device for digital image capture can be embodied as a digital camera, a smartphone, smart glasses, or another electronic device with a digital image capture function. The device for digital image output can be embodied as the device for digital image capture or, alternatively, as a virtual reality headset or an augmented reality headset. The digital camera device can be part of a smartphone or a dedicated digital camera. The image generation unit can be embodied at least partially or entirely as an optical detector unit, in particular a camera sensor unit. The optical detector unit can be a camera sensor unit with two lenses and / or with two photo sensors.The optical detector unit can also be a camera with a lens assigned to a photosensor, wherein the sensor switches back and forth between two different recording modes to generate two differently parameterized images. The different recording modes can be different camera settings, e.g. with regard to zoom, brightness, sharpness, depth of focus or other known parameters that can be set in a camera. Alternatively or additionally, the image generation unit can be designed at least partially or completely as a digital image generation device which, in particular, does not generate the image by reproducing an image motif that actually exists in an environment of the device, but instead generates it purely virtually, e.g. as an augmented reality object or as a virtual reality environment. The digital image generation device is, for example, a computing unit which generates artificial images.In particular, the digital image generation device can generate the artificial images itself for the first time or receive them from external sources and make them available for image synthesis.

[0006] In particular, the scene can correspond completely or partially to a photographic image of a real environment of the device and / or be completely or partially digitally generated. “Anisotropic image synthesis” should be understood in particular as image synthesis in which individual images are combined with one another, whereby the individual images are changed in different directions, e.g. compressed, distorted, zoomed, etc. In the aforementioned example of a mountain range, the mountain range could, in anisotropic image synthesis, completely fill a vertical image axis and be compressed along a horizontal image axis so that the mountain range can be depicted across its entire width - or vice versa, a flat mountain range could be stretched upwards - while optionally, due to a context dependency, the person in the foreground remains unchanged, i.e. in particular not compressed laterally orwould be stretched upwards. Alternatively or additionally, content-dependent, in particular isotropic, digital image synthesis is conceivable, in which a background is changed isotropically, i.e. without changes in proportion, e.g. zoomed, while the person in the foreground remains unchanged. For example, a moon in a night sky could be zoomed in separately from the rest of the image. Optional additional aesthetic image processing using the digital image processing device is conceivable, e.g. to aesthetically conceal any distortions that may arise. The digital image processing device can be embodied by a computing unit. A “computing unit” is to be understood in particular as a unit with an information input, information processing and information output.Advantageously, the computing unit comprises at least one processor, a memory, input and output means, additional electrical components, an operating program, control routines, and / or calculation routines. The computing unit can be integrated into the digital camera or smartphone or external to it, e.g., as a cloud connected to the digital camera or smartphone.

[0007] It is further proposed that the different parameterization of the at least two images of the scene, in particular a completely real, completely artificial, or partially artificial and partially real, each comprise different perspectives, different fields of view, and / or different digital and / or analog zoom factors (digitally and / or analogically adjusted focal lengths). This advantageously makes it possible to overcome the limitations of conventional photography. The novel functionality can advantageously be created for image recording devices. Image recording and / or image output can advantageously be improved. Preferably, the image generation unit is provided to record the two images of the scene with different analog and / or digital zoom and / or different digitally generated or analogically adjusted focal length settings of the camera.“Intended” and / or “configured” should be understood in particular to mean specially programmed, designed and / or equipped. The fact that an object is intended or configured for a specific function should be understood in particular to mean that the object fulfills and / or performs this specific function in at least one application and / or operating state. In particular, fully or partially digitally generated images can comprise different perspectives, fields of view or zoom factors or at least image components with different perspectives, fields of view or zoom factors. In particular, fully or partially photographic images of a real environment can be created with different camera settings, such as zooms, focal lengths, depth of focus, sharpness settings, etc.

[0008] It is further proposed that the digital image processing device be configured to perform an interpolation and / or an extrapolation between the at least two differently parameterized images when generating the anisotropic digital image synthesis by combining the at least two images. This advantageously allows a particularly high quality of the image generated by the digital image synthesis to be achieved. A particularly realistic image impression can be obtained.

[0009] In addition, it is proposed that the digital image processing device for anisotropic digital image synthesis, in particular by means of interpolation and / or extrapolation, be configured to perform, in particular set, dimensioning of the scene or a component from the at least two images used to create the combined image along at least one directional axis of an image coordinate system of the combined image, in particular along two mutually independent, preferably mutually orthogonal, directional axes of the image coordinate system, taking into account a, in particular adjustable, weighting ratio of the at least two images. This advantageously makes it possible to provide a particularly diverse and / or flexible image processing function. In particular, at least one of the directional axes, preferably the directional axis, is a horizontal line.Alternatively, the directional axis can also run at an angle to the horizontal. In particular, at least one of the directional axes, preferably a further directional axis different from the directional axis, is vertical. Alternatively, the further directional axis can also run at an angle to the vertical. For example, in a combined image generated from two images each depicting the same scene or the same component with different parameterization, e.g., different zoom, depending on the weighting of the individual images, the dimensions of the scene or component are closer to the dimensions from one image or the other image.For example, if the scene or component in one of the two images is assigned a high weighting and not in the other image, then the dimensions of the scene or component in the combined image will be close to the dimensions of the scene or component in the image with the high weighting. Thus, in the aforementioned example with the mountain range, if a high vertical weighting is set for the image with the higher zoom, the mountain range will appear higher in the combined image than in a photograph created using the conventional method.

[0010] It is further proposed that the device comprise a position detection unit which is at least configured to determine a relationship of the image coordinate system to a position of at least the image generation unit in a physical environment of the image generation unit, preferably the camera sensor. This advantageously simplifies operation. Furthermore, simple weighting along horizontal and vertical axes can advantageously be enabled. The position detection unit can be designed, for example, as a gyro sensor, as an acceleration sensor, as an inertial sensor, as a magnetometer or as another type of orientation sensor. The position detection unit preferably determines a relative position of the directional axis to the horizontal. The position determined by the position detection unit can be output to the user, e.g. as a horizon line on a display of the device.Alternatively or additionally, the position determined by the position detection unit can serve as input for the software of the digital image processing system. In this case, an explicit display as a horizontal line could be omitted.

[0011] Furthermore, it is proposed that the digital image processing device is configured, at least for the content-dependent, in particular isotropic or anisotropic, digital image synthesis from the at least two images of the scene, in particular by means of interpolation and / or extrapolation, to carry out, in particular to set, a context-dependent dimensioning of components within the scene depicted in the combined image, taking into account a relative weighting of the at least two differently parameterized images. This advantageously makes it possible to provide a particularly diverse and / or flexible image processing function. Advantageously, a component of a scene, for example a person, can be processed, e.g., enlarged or reduced, independently of the rest of the scene, for example the background.For example, in a combined image created from two images each depicting the same scene with the same component but with different parameterization, e.g. different zoom, the dimensioning of the component will be closer to the dimensioning of the components in one image or the other, depending on the weighting of the individual images. For example, if a component in one of the two images is given a high weighting but not in the other, the dimensioning of the component in the combined image will be close to the dimensioning of the component in the image with the high weighting. So, in the above example with the mountain range and the person, if a weighting is set on the "person" component for the image with the higher zoom, the person will appear larger in the combined image than in a photograph created using the classic method.In particular, for context-dependent component dimensioning, the context of the component may be a distance of the component from the image generation unit or an entity of the component, for example, a subject such as the person in contrast to a background such as a landscape or a sky, etc.

[0012] Furthermore, it is proposed that the digital image processing device be configured at least to recognize components and / or contexts of components within the images generated by the image generation unit. This advantageously allows for simple operation. Furthermore, a diverse and / or flexible image processing function can advantageously be provided. In particular, the recognition of the components and / or contexts of components can be performed using known image processing methods, e.g., object recognition or the like. In particular, the digital image processing device has an object recognition algorithm for this purpose.

[0013] Furthermore, it is proposed that the device has a user interface which provides at least one weighting setting function for setting and / or adjusting the relative weighting of the at least two differently parameterized images, at least one image coordinate system setting function for at least setting and / or adjusting an orientation of the image coordinate system and / or at least one component selection function for at least selecting the component for the context-dependent dimensioning. This can advantageously enable user-friendly, simple and / or intuitive operation of the device. For example, the weighting setting function could be operated via a touchscreen, e.g. using two-finger zoom gestures (“pinch-to-zoom”). For example, the image coordinate system setting function could be operated via a touchscreen, e.g.The component selection function can be operated using two-finger zoom gestures and / or two-finger rotation gestures. In particular, the extent and coordinate system of the anisotropic image synthesis, in particular the anisotropic zoom, can be controlled via the extent and direction of a finger movement. For example, the component selection function could be operated via a touchscreen, e.g., by tapping. In particular, the context for the content-dependent image synthesis, preferably the context-dependent dimensioning, can be defined by clicking on a component of a scene with structure recognition.

[0014] If the digital image processing device is configured to carry out the digital image synthesis, in particular an interpolation and / or extrapolation of the at least two images, a component recognition from the images generated by the image generation unit and / or a context recognition of the components, with the aid of trainable and / or trained algorithms, preferably machine learning algorithms, and / or if the digital image processing device is configured to automatically correct image artifacts arising during the digital image synthesis, for example distortions or superimpositions, a high quality and / or speed of the image processing can advantageously be achieved.

[0015] Furthermore, the digital image processing device is proposed for the device for digital image recording and / or digital image output and / or a digital camera with the device for digital image recording and / or digital image output. The digital camera can be designed as a standalone digital camera or as a digital camera integrated into an electronic device with additional functions, such as a smartphone.

[0016] Furthermore, a method for digital image capture, preferably for a digital camera, is proposed, wherein in a generation step at least two differently parameterized images of a scene, in particular a completely real, completely artificial or partially artificial and partially real scene, are generated, in particular recorded, and wherein in a digital image processing step, in which the at least two generated, in particular recorded, differently parameterized images are combined by means of anisotropic and / or content-dependent digital image synthesis of the at least two images, and thereby a combined image, which is at least partially modified in comparison to the original scene, is generated and preferably output to a user. This advantageously creates a novel functionality for image capture devices, such as digital cameras or smartphones.Image capture and / or image output can be advantageously improved. The number of options for image processing and / or image optimization, particularly directly on image capture devices, can be advantageously increased.

[0017] The device according to the invention, the digital image processing device according to the invention, the digital camera according to the invention, and the method according to the invention are not intended to be limited to the application and embodiment described above. In particular, the device according to the invention, the digital image processing device according to the invention, the digital camera according to the invention, and the method according to the invention can have a number of individual elements, components, units, and method steps that differs from the number stated herein in order to fulfill a functionality described herein. Furthermore, in the value ranges specified in this disclosure, values ​​within the stated limits are also to be considered disclosed and can be used arbitrarily. drawing

[0018] Further advantages will become apparent from the following description of the drawings. The drawings illustrate an exemplary embodiment of the invention. The drawings, the description, and the claims contain numerous features in combination. Those skilled in the art will also expediently consider the features individually and combine them into useful further combinations.

[0019] They show: Fig. 1 schematically shows an exemplary smartphone with a digital camera having a device according to the invention for digital image recording and / or digital image output, Fig. 2a an exemplary first image of a scene created by means of the device, Fig. 2b an exemplary second image of the scene created by the device, Fig. 2c an exemplary output of a user interface of the device, with a first exemplary representation of an anisotropic image synthesis of the two images containing the scene, Fig. 2d an exemplary output of a user interface of the device, with a second exemplary representation of an anisotropic image synthesis of the two images containing the scene, Fig. 3a shows an exemplary first image of another scene created by means of the device, Fig. 3b an exemplary second image of the further scene created by means of the device, Fig. 3c an exemplary output of a user interface of the device, with a third exemplary representation of an anisotropic image synthesis of the two images containing the further scene, Fig. 3d an exemplary output of a user interface of the device, with a fourth exemplary representation of an anisotropic image synthesis of the two images containing the further scene, Fig. 4a shows an exemplary first image created by the device, which comprises two components, Fig. 4b shows an exemplary second image created by the device, which includes the two components, Fig. 4c shows an exemplary output of a user interface of the device, with a fifth exemplary representation of an anisotropic image synthesis of the two images containing the two components, Fig. 4d an exemplary output of a user interface of the device, with a sixth exemplary representation of an anisotropic image synthesis of the two images containing the two components and Fig. 5 a schematic flow diagram of a method for digital image acquisition. Description of the embodiment

[0020] The Fig. 1 shows, by way of example, a smartphone with a digital camera 42. Alternatively, the digital camera 42 could also be part of a standalone camera or another electrical device. The digital camera 42 comprises a device 40 for digital image recording and / or digital image output. The device 40 has an image generation unit 10. The image generation unit 10 is configured to record and generate at least two differently parameterized Fig. (cf. Fig. 2a, Fig. 2b, Fig. 3a, Fig. 3b, Fig. 4a, Fig. 4b) of a scene 16, 18 (cf. Fig. 2a to 4d). The image generation unit 10 is designed as an optical detector unit in the example shown. The optical detector unit has a plurality of separate photosensors 48, 50 in the example shown. A first photosensor 48 is configured to record the first Fig. with the first parameterization. A second photosensor 50 is configured to record the second Fig. with the second parameterization. The different parameterization of the two Fig. Scene 16, 18 includes different zoom factors (cf. among others Fig. 2a and Fig. 2b). Alternatively or additionally, the different parameterization of the two Fig. also include different fields of view and / or different perspectives. Alternatively or additionally, the image generation unit 10 could be designed as a digital image generation unit (not shown) that Fig. artificially produced.

[0021] The device 40 comprises a digital image processing device 20. The digital image processing device 20 is at least configured to process the at least two recorded and differently parameterized Fig. to be combined in such a way that an anisotropic digital image synthesis (cf. Fig. 2c, Fig. 2d, Fig. 3c, Fig. 3D, Fig. 4c and Fig. 4d) and / or by content-dependent digital image synthesis (cf. Fig. 4c and Fig. 4d) of at least two Fig. , a combined image 22 is generated that is modified compared to the original scene 16, 18. The digital image processing device 20 is configured to automatically correct image artifacts arising during digital image synthesis, such as distortions or overlays. The device 40 has a user interface 32. The user interface 32 is embodied, for example, as a display of the smartphone. The user interface 32 is configured to visually output the combined image 22 to a user.

[0022] The device 40 has a position detection unit 56. The position detection unit 56 is configured to determine a spatial position of the image generation unit 10. The position detection unit 56 is formed by a position sensor of the smartphone. The position detection unit 56 is provided to determine a reference of an image coordinate system of the Fig. and / or the combined image 22 to a position of the image generation unit 10 in a physical environment of the image generation unit 10.

[0023] The Fig. 2a to 4d each show exemplary visual outputs that the user receives via the user interface 32. The Fig. 2a shows an example of the first (unprocessed) Fig. a scene 16, which is shown here in simplified form as a triangle. The first Fig. is recorded with a first focal length / with a first zoom factor by the image generation unit 10. The Fig. 2b shows an example of the second (unprocessed) Fig. the same scene 16 as in the Fig. 2a. The second Fig. is recorded with a second focal length / with a second zoom factor by the image generation unit 10. For example, the first zoom factor is larger than the second zoom factor. Fig. 2c and Fig. 2d show exemplary combined images 22, which in the case shown here are created by the anisotropic digital image synthesis of the two Fig. from the Fig. 2a and Fig. 2b. The combined image 22 is compared to the two Fig. modified (here: compared to the Fig. 2a horizontally compressed and compared to the Fig. 2b stretched vertically). The digital image processing device 20 is designed to generate the anisotropic digital image synthesis by combining the two Fig. , an interpolation and / or an extrapolation between the two differently parameterized Fig. The digital image processing device 20 is designed to carry out the digital image synthesis, in particular the interpolation and / or extrapolation of the two Fig. with the aid of trainable and / or trained algorithms, preferably machine learning algorithms. The digital image processing device 20 is configured for the anisotropic digital image synthesis by means of interpolation and / or extrapolation to determine the dimensioning of the scene 16 or a component 28, 30 (see Fig. 4a to 4d) from the at least two images used to create the combined image 22 Fig. along a directional axis 24 of an image coordinate system of the combined image 22 or along two mutually independent, preferably mutually orthogonal, directional axes 24, 26 of the image coordinate system, taking into account a, in particular adjustable, weighting ratio of the two Fig. to undertake, in particular to discontinue. Fig. 2c and Fig. 2d show different weighting ratios. In the Fig. 2c is an example of the weighting of the second Fig. in the horizontal direction significantly greater than the weighting of the first Fig. . In the Fig. 2c is an example of the weighting of the second Fig. in the vertical direction significantly smaller than the weighting of the first Fig. . In the Fig. 2d, the weighting ratio in the vertical direction is approximately the same as in the Fig. 2c, but in the horizontal direction the weighting ratio is also in the direction of the first Fig. delay.

[0024] The Fig. 2c and Fig. 2d each show representations as they could be presented to a user via the user interface 32. In this case, the user interface 32 provides a weighting adjustment function 34 for setting and / or adjusting the relative weighting of the two differently parameterized Fig. The user can then adjust the weighting ratio along the corresponding directional axis 24, for example, using a two-finger zoom gesture along the dashed arrows.

[0025] The Fig. 3a shows an example of the first (unprocessed) Fig. another scene 18, which is shown here in simplified form as a circle. The first Fig. is recorded with a first focal length / with a first zoom factor by the image generation unit 10. The Fig. 3b shows an example of the second (unprocessed) Fig. the same further scene 18 as in the Fig. 3a. The second Fig. is recorded with a second focal length / with a second zoom factor by the image generation unit 10. For example, the first zoom factor is larger than the second zoom factor. Fig. 3c and Fig. 3d show exemplary combined images 22, which in the case shown here are created by the anisotropic digital image synthesis of the two Fig. from the Fig. 3a and Fig. 3b. The combined image 22 is compared to the two Fig. modified (here: compared to the Fig. 3a and Fig. 3b compressed and stretched diagonally to the horizontal). In the example of Fig. 3a to 3d, the directional axes 24, 26 differ in contrast to the example of Fig. 2a to 2d from a horizontal axis 52 and a vertical axis 54. The directional axes 24, 26, along which the weighting of the Fig. to each other, are thus inclined / rotated relative to the horizontal axis 52 or the vertical axis 54.

[0026] The Fig. 3c and Fig. 3d each show representations as they could be presented to a user via the user interface 32. In this case, the user interface 32 provides an image coordinate system setting function 36 for setting and / or adjusting an orientation of the image coordinate system. The user can then, for example, adjust the rotational orientation of the image coordinate system, in particular of the directional axes 24, 26, by rotating the fingers on the touchpad. Subsequently, the user can then adjust the weighting ratio along the correspondingly rotated directional axes 24, 26 using the weighting setting function 34 by a two-finger zoom gesture along the dashed arrows. The real horizontal axis 52 is available to the device 40 via the position detection unit 56. This can - as in the example of the Fig. 3d - can be output to the user via the user interface 32, e.g. as an indicated line, e.g. to support operation of the image coordinate system setting function 36.

[0027] The Fig. Figure 4a shows an example of a first (unprocessed) Fig. , which contains two components 28, 30, which are simplified here as the circle and the triangle. The first Fig. is recorded with a first focal length / with a first zoom factor by the image generation unit 10. The Fig. 4b shows an example of a second (unprocessed) Fig. , which also contain the same components 28, 30 as the Fig. 3a contains. The second Fig. is recorded with a second focal length / with a second zoom factor by the image generation unit 10. For example, the first zoom factor is greater than the second zoom factor. The digital image processing device 20 is at least for the content-dependent digital image synthesis from the at least two Fig. of the scene 16, 18 by means of interpolation and / or extrapolation to determine a context-dependent dimensioning of the components 28, 30 within the scene 16, 18 depicted in the combined image 22, taking into account a relative weighting of the at least two differently parameterized Fig. to undertake, in particular to discontinue. Fig. 4c and Fig. 4d show exemplary combined images 22, which in the case shown here are created by a content-dependent and anisotropic image synthesis of the two Fig. from the Fig. 4a and Fig. 4b. The combined image 22 is compared to the two Fig. modified. In the Fig. 4a is the first component 28 compared to the first Fig. horizontally compressed (ie the first Fig. is weighted lower in the horizontal direction) and compared to the second Fig. inserted vertically (ie the first Fig. is weighted higher in the vertical direction). At the same time, the second component 30 is approximately identical to its representation in the first Fig. . The two components 28, 30 are thus modified independently of each other. The two components 28, 30 thus flow differently into the image synthesis. The first component 28 (triangle) can represent a distant object or background (e.g., the mountain range), and the second component 30 (circle) a nearby object or motif (e.g., the person). The two Fig. These scenes 16, 18 are combined to form the content-dependent synthesis in such a way that only the first component 28 is changed anisotropically, while the second component 30 is completely assigned to one of the two Fig. (here: Fig. ). In one application, this behavior could mean that a background (e.g., a mountain range) is compressed horizontally while the subject remains unchanged. This advantageously allows the relevant subject to remain unchanged, while, for example, more landscape, where a horizontal compression is less noticeable, can be displayed.

[0028] The Fig. 4c and Fig. 4d each show representations as they could be presented to a user via the user interface 32. The user interface 32 provides, as an example, the weighting adjustment function 34 for only one of the components 28, 30. In addition, the user interface 32 provides a component selection function 38 (see Fig. 4d) to a selection of the active component 28, 30 for the context-dependent dimensioning, e.g., by means of the weighting setting function 34 and / or by means of the image coordinate system setting function 36. The user can, for example, select which of the components 28, 30 is currently active by tapping one of the components 28, 30. The currently active component 28, 30 is then the one that can currently be modified by the user by applying the weighting setting function 34 and / or the image coordinate system setting function 36. It is conceivable that the content-dependent image synthesis refers by default to a preselected component 28, 30, preferably the component 28 comprising the background. The user can then select a different component 28, 30 by means of the component selection function 38. In the Fig. Figure 4d shows a possible implementation in which a click on the first component 28 (on the triangle) marks it as relevant, so that the following anisotropic image synthesis, for example, only affects the other component 30 and the relevant component 28, 30 is kept unchanged relative to the current representation (which may or may not already be anisotropically modified).

[0029] The digital image processing device 20 is configured to identify components 28, 30 and / or contexts of components 28, 30 within the image generated by the image generation unit 10. Fig. The digital image processing device 20 is configured to automatically detect the components from the images generated by the image generation unit 10. Fig. and / or to carry out the context recognition of the components 28, 30 with the aid of trainable and / or trained algorithms, preferably machine learning algorithms.

[0030] The Fig. Figure 5 shows a schematic flow diagram of a method for digital image acquisition. The method is intended at least for use in the digital camera 42. In a generation step 44, at least the two differently parameterized Fig. of the scene 16, 18 is generated, in particular recorded. In the generation step 44, a component recognition and / or a context recognition can be carried out. In a digital image processing step 46, the at least two generated, in particular recorded, differently parameterized Fig. by means of anisotropic and / or content-dependent digital image synthesis. In the digital image processing step 46, the combined image 22 is thereby generated. The combined image 22 is generated as an image 22 that is at least partially modified compared to the original scene 16, 18 and is output to the user via the user interface 32. In the anisotropic and / or content-dependent digital image synthesis, user inputs made by the user via the weighting setting function 34, via the image coordinate system setting function 36 and / or via the component selection function 38 are taken into account and implemented. In addition, in the anisotropic and / or content-dependent digital image synthesis, a position of the components 28, 30 determined by the position detection unit 56 within the Fig.be taken into account. The digital image processing step 46 may also include further image processing steps, such as further aesthetic distortion corrections or the like.

Claims

[1] Device (40) for digital image recording and / or digital image output, in particular a digital camera device, with at least one image generation unit (10) which is at least designed to generate, in particular to record, at least two differently parameterized images (12, 14) of a scene (16, 18), in particular a completely real, completely artificial or partially artificial and partially real scene, characterized bya digital image processing device (20) which is at least configured to combine the at least two generated, in particular recorded, differently parameterized images (12, 14) with one another in such a way that a combined image (22), which is at least partially modified in particular compared to the original scene (16, 18), is generated by anisotropic and / or content-dependent digital image synthesis of the at least two images (12, 14) and is preferably output to a user. [2] Device (40) according to claim 1, characterized by that the different parameterization of the at least two images (12, 14) of the scene (16, 18) each comprises different perspectives, different fields of view and / or different digital and / or analog zoom factors. [3] Device (40) according to claim 1 or 2, characterized byin that the digital image processing device (20) is designed to carry out an interpolation and / or an extrapolation between the at least two differently parameterized images (12, 14) when generating the anisotropic digital image synthesis by combining the at least two images (12, 14). [4] Device (40) according to claim 3, characterized byin that the digital image processing device (20) for the anisotropic digital image synthesis, in particular by means of interpolation and / or extrapolation, is set up to carry out, in particular to set, a dimensioning of the scene (16, 18) or of a component (28, 30) from the at least two images (12, 14) used to create the combined image (22) along at least one directional axis (24, 26) of an image coordinate system of the combined image (22), in particular along two mutually independent, preferably mutually orthogonal, directional axes (24, 26) of the image coordinate system, taking into account a, in particular adjustable, weighting ratio of the at least two images (12, 14). [5] Device (40) according to claim 4, characterized bya position detection unit (56) which is at least configured to determine a reference of the image coordinate system to a position of at least the image generation unit (10) in a physical environment of the image generation unit (10). [6] Device (40) according to one of claims 3 to 5, characterized by that the digital image processing device (20) is designed, at least for the content-dependent digital image synthesis from the at least two images (12, 14) of the scene (16, 18), in particular by means of interpolation and / or extrapolation, to carry out, in particular to set, a context-dependent dimensioning of components (28, 30) within the scene (16, 18) depicted in the combined image (22), taking into account a relative weighting of the at least two differently parameterized images (12, 14). [7] Device (40) according to claim 6, characterized bythat the digital image processing device (20) is at least configured to recognize components (28, 30) and / or contexts of components (28, 30) within the images (12, 14) generated by the image generation unit (10). [8] Device (40) according to at least one of claims 4 or 6, characterized by a user interface (32) which provides at least one weighting setting function (34) for setting and / or adapting the relative weighting of the at least two differently parameterized images (12, 14) and / or at least one image coordinate system setting function (36) at least for setting and / or adapting an orientation of the image coordinate system and / or at least one component selection function (38) at least for selecting the component (28, 30) for the context-dependent dimensioning. [9] Device (40) according to one of the preceding claims, characterized byin that the digital image processing device (20) is designed to carry out the digital image synthesis, in particular an interpolation and / or extrapolation of the at least two images (12, 14), a component recognition from the images (12, 14) generated by the image generation unit (10) and / or a context recognition of the components (28, 30), with the aid of trainable and / or trained algorithms, preferably machine learning algorithms. [10] Device (40) according to one of the preceding claims, characterized by that the digital image processing device (20) is designed to automatically correct image artifacts arising during digital image synthesis, for example distortions or superimpositions. [11] Digital image processing device (20) for a device (40) for digital image recording and / or for digital image output according to one of the preceding claims. [12] Digital camera (42) with a device (40) for digital image recording and / or for digital image output according to one of claims 1 to 10. [13] Method for digital image recording, preferably for a digital camera (42) according to claim 12, wherein in a generation step (44) at least two differently parameterized images (12, 14) of a scene (16, 18), in particular a completely real, completely artificial or partially artificial and partially real scene, are generated, in particular recorded, characterized bya digital image processing step (46) in which the at least two generated, in particular recorded, differently parameterized images (12, 14) are combined by means of an anisotropic and / or content-dependent digital image synthesis of the at least two images (12, 14) and thereby a combined image (22) is generated, which is at least partially modified in particular compared to the original scene (16, 18), and is preferably output to a user.

Images