TW201913562A - Device and method for generating panorama image - Google Patents

Device and method for generating panorama image Download PDF

Info

Publication number
TW201913562A
TW201913562A TW106129172A TW106129172A TW201913562A TW 201913562 A TW201913562 A TW 201913562A TW 106129172 A TW106129172 A TW 106129172A TW 106129172 A TW106129172 A TW 106129172A TW 201913562 A TW201913562 A TW 201913562A
Authority
TW
Taiwan
Prior art keywords
image
panoramic
module
panoramic image
images
Prior art date
Application number
TW106129172A
Other languages
Chinese (zh)
Other versions
TWI636424B (en
Inventor
邱冠榮
陳祐誠
何其勳
Original Assignee
鴻海精密工業股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 鴻海精密工業股份有限公司 filed Critical 鴻海精密工業股份有限公司
Priority to TW106129172A priority Critical patent/TWI636424B/en
Application granted granted Critical
Publication of TWI636424B publication Critical patent/TWI636424B/en
Publication of TW201913562A publication Critical patent/TW201913562A/en

Links

Abstract

A panorama image generating device includes an image library, a receiving module, an analysis module, a first extracting module, a selecting module, a second extracting module, and a splicing module. The image library stores a plurality of images. The receiving module is configured to receive basic searching information. The analysis module is configured to obtain key searching information of the required image according to the basic searching information. The first extracting module is configured to exact first images which are matching the key searching information from the image library. The selecting module is configured to select second images which comprise different azimuth angles and match predetermined environmental parameter from the first images. The second extracting module is configured to extract optimization images of each azimuth angle from the second images according to an environmental parameter and an image parameter of each second image. The splicing module is configured to regulate the image parameter of each optimization image to splice a panorama image. A panorama image generating method is also provided.

Description

全景影像生成裝置及方法Panoramic image generating device and method

本發明涉及影像處理技術領域,尤其涉及一種全景影像生成裝置及方法。The present invention relates to the field of image processing technologies, and in particular, to a panoramic image generating device and method.

隨著智慧手機和數碼相機的興起,透過圖像記錄生活內容已經是普遍現象。線上雲端與線下終端儲存著使用者海量的圖像, 其中絕多數為2D圖像, 這些圖像資料報含了許多當時拍攝的環境參數。而全景影像可以讓無法親臨現場或想再次回味當時景像的用戶身臨其境,如何將這些拍攝散落在各地的圖像資訊作有效率且實用的篩選、整合, 進而將其彙整生成各種角度的全景影像是一大技術挑戰。With the rise of smart phones and digital cameras, it has become common to record life content through images. The online cloud and offline terminals store a large number of users' images, most of which are 2D images. These image data reports contain many environmental parameters taken at that time. The panoramic image can make the users who can't visit the scene or want to relive the scene at the scene immersively, how to effectively and practically filter and integrate the image information scattered around the scene, and then integrate it to generate various angles. The panoramic image is a major technical challenge.

有鑑於此,有必要提供一種全景影像生成裝置及方法,其能將2D圖像進行拼接生成全景影像。In view of the above, it is necessary to provide a panoramic image generating device and method capable of splicing 2D images to generate panoramic images.

本發明一實施方式提供一種全景影像生成裝置,包括一圖像資料庫,所述圖像資料庫存儲有多張圖像。所述全景影像生成裝置還包括:接收模組,用於接收一基礎檢索資訊;解析模組,用於根據所述基礎檢索資訊解析出所需檢索的圖像資訊;第一提取模組,用於從所述圖像資料庫中提取出符合所述圖像資訊的第一圖像;篩選模組,用於從所述提取的第一圖像中篩選出具有不同方位角度且符合預設環境參數的第二圖像;第二提取模組,用於根據所述每一第二圖像的環境參數與圖像參數從中提取出各方位角度的優選圖像;及拼接生成模組,用於調整所述每一優選圖像的圖像參數以進行拼接生成一全景影像。An embodiment of the present invention provides a panoramic image generating apparatus, including an image database, wherein the image data library stores a plurality of images. The panoramic image generating device further includes: a receiving module, configured to receive a basic search information; and an analyzing module, configured to parse the image information that needs to be retrieved according to the basic search information; Extracting, from the image database, a first image that meets the image information; and a screening module, configured to filter out the first image from the extracted image and have a different azimuth angle and conform to a preset environment a second image of the parameter; a second extraction module, configured to extract a preferred image of each of the angles according to the environmental parameters and the image parameters of each of the second images; and a mosaic generation module, configured to: The image parameters of each of the preferred images are adjusted for splicing to generate a panoramic image.

本發明一實施方式提供一種全景影像生成方法,包括以下步驟:接收一基礎檢索資訊;根據所述基礎檢索資訊解析出所需檢索的圖像資訊;從圖像資料庫中提取出符合所述圖像資訊的第一圖像;從所述提取的第一圖像中篩選出具有不同方位角度且符合預設環境參數的第二圖像;根據所述每一第二圖像的環境參數與圖像參數從中提取出各方位角度的優選圖像;及調整所述每一優選圖像的圖像參數以進行拼接生成一全景影像。An embodiment of the present invention provides a method for generating a panoramic image, comprising the steps of: receiving a basic search information; parsing the image information that is required to be retrieved according to the basic search information; and extracting the image from the image database. a first image of the information; a second image having different azimuth angles and conforming to the preset environmental parameter is filtered out from the extracted first image; and environmental parameters and graphs according to each of the second images A preferred image is extracted from the angle of each of the pixels, and the image parameters of each of the preferred images are adjusted to be stitched to generate a panoramic image.

與現有技術相比,上述全景影像生成裝置及方法,可以根據接收到的基礎檢索資訊(語音/文字/圖像)來篩選圖像並將圖像進行拼接生成全景影像供使用者觀看,提示用戶的觀看體驗。Compared with the prior art, the above-mentioned panoramic image generating device and method can filter images according to the received basic search information (speech/text/image) and splicing the images to generate a panoramic image for the user to view, prompting the user Viewing experience.

請參閱圖1-2,在一實施方式中,一種全景影像生成裝置100用於實現將不同角度拍攝的圖像進行拼接形成一預設角度的全景影像。例如可以是180度全景影像、270度全景影像、360度全景影像。其中,180度全景影像優選是一個半球形的全景影像,360度全景影像優選是一個球形的全景影像。Referring to FIG. 1-2, in an embodiment, a panoramic image generating apparatus 100 is configured to splicing images captured at different angles to form a panoramic image of a preset angle. For example, it can be a 180 degree panoramic image, a 270 degree panoramic image, or a 360 degree panoramic image. The 180-degree panoramic image is preferably a hemispherical panoramic image, and the 360-degree panoramic image is preferably a spherical panoramic image.

全景影像生成裝置100包括圖像資料庫10、接收模組20、解析模組30、第一提取模組40、篩選模組50、第二提取模組60及拼接生成模組70。The panoramic image generating apparatus 100 includes an image database 10, a receiving module 20, an analysis module 30, a first extraction module 40, a screening module 50, a second extraction module 60, and a splicing generation module 70.

圖像資料庫10存儲有多張圖像。圖像資料庫10也可以設置在雲端伺服器(圖未示),全景影像生成裝置100通過訪問雲端伺服器來連結至圖像資料庫10。當然,圖像資料庫10也可以是全景影像生成裝置100內存儲的圖像與雲端伺服器存儲的圖像的集合。每一圖像可以包括與其相關的環境參數及圖像參數。環境參數可以是:拍攝時的方位角度、光圈、快門速度、曝光值、光線角度、風速、溫濕度中的一種或多種。圖像參數可以是:景深資訊、顏色、明亮度、銳利度、焦距中的一種或多種。The image database 10 stores a plurality of images. The image database 10 may also be installed in a cloud server (not shown), and the panoramic image generating device 100 is connected to the image database 10 by accessing a cloud server. Of course, the image database 10 may be a collection of images stored in the panoramic image generating device 100 and images stored in the cloud server. Each image may include environmental parameters and image parameters associated therewith. The environmental parameter may be one or more of azimuth angle, aperture, shutter speed, exposure value, light angle, wind speed, temperature and humidity at the time of shooting. The image parameters may be one or more of depth of field information, color, brightness, sharpness, and focal length.

接收模組20用於接收一基礎檢索資訊。所述基礎檢索資訊可以包括圖像、語音資訊、文字資訊中的一種或多種。舉例而言,在2014年阿里山櫻花季開幕日(2014年3月10日),某一用戶無法親臨現場或想再次回味當時的景象。該使用者可以以語音或文字形式輸入“2014年阿里山櫻花季開幕日”資訊至全景影像生成裝置100,全景影像生成裝置100即可獲知需要將屬於“2014年阿里山櫻花季開幕日”的圖像進行拼接形成一預設角度的全景影像供該使用者以各種裝置進行觀看(可以是移動設備、PC端、VR設備,也可以是全景影像生成裝置100)。用戶也可以上傳至少一張屬於“2014年阿里山櫻花季開幕日”的圖像至全景影像生成裝置100來作為輸入的基礎檢索資訊。The receiving module 20 is configured to receive a basic search information. The basic search information may include one or more of image, voice information, and text information. For example, in the opening day of the Alishan Cherry Blossom Season in 2014 (March 10, 2014), a user could not visit the scene or wanted to relive the scene at that time. The user can input the "2014 Alishan Cherry Blossom Season Opening Day" information to the panoramic image generating device 100 in the form of voice or text, and the panoramic image generating device 100 can know that it is necessary to belong to the "2014 Alishan Cherry Blossom Season Opening Day". The images are spliced to form a panoramic image of a predetermined angle for the user to view with various devices (either a mobile device, a PC end, a VR device, or a panoramic image generating device 100). The user can also upload at least one image belonging to the "2014 Alishan Cherry Blossom Season Opening Day" to the panoramic image generating device 100 as the basic search information for input.

解析模組30用於根據所述基礎檢索資訊解析出所需檢索的圖像資訊。舉例而言,若接收模組20接收到使用者以語音或文字形式輸入“2014年阿里山櫻花季開幕日”的基礎檢索資訊。解析模組30經過語意分析得出欲檢索的圖像資訊包括:時間(2014年)、地點(阿里山櫻花季現場)及精確時間區間(開幕日、3月10日)。若接收模組20接收到使用者輸入的是一張圖像。解析模組30通過圖像分析得出此圖像的拍攝時間是2014年3月10日,地點是阿里山櫻花季現場,影像內有吊橋、入口門牌、櫻花等。解析模組30經過對使用者輸入的圖像進行解析同樣可得出欲檢索的圖像資訊包括:時間(2014年)、地點(阿里山櫻花季現場)及精確時間區間(開幕日、3月10日)。The parsing module 30 is configured to parse the image information that needs to be retrieved according to the basic search information. For example, if the receiving module 20 receives the basic search information of the "2014 Alishan Cherry Blossom Season Opening Day" in the form of voice or text. The parsing module 30 analyzes the semantics to obtain the image information to be retrieved including: time (2014), location (Alishan cherry blossom season scene) and precise time interval (opening day, March 10). If the receiving module 20 receives an input from the user, it is an image. The analysis module 30 analyzes the image by the image analysis on March 10, 2014. The location is the scene of the Alishan cherry blossom season. The image includes a suspension bridge, an entrance door, and cherry blossoms. The analysis module 30 can also obtain the image information to be retrieved by analyzing the image input by the user, including: time (2014), location (Alishan cherry blossom season scene) and precise time interval (opening day, March) 10th).

第一提取模組40用於從圖像資料庫10中提取出符合所述圖像資訊的第一圖像。舉例而言,若圖像資料庫10包含有200張屬於“2014年阿里山櫻花季開幕日”拍攝的圖片,第一提取模組40從圖像資料庫10提取出該200張圖像(即200張第一圖像)。The first extraction module 40 is configured to extract, from the image database 10, a first image that conforms to the image information. For example, if the image database 10 contains 200 pictures taken in the "Alishan Cherry Blossom Season Opening Day 2014", the first extraction module 40 extracts the 200 images from the image database 10 (ie, 200 first images).

篩選模組50用於從所述提取的第一圖像中篩選出具有不同方位角度且符合預設環境參數的第二圖像。舉例而言,所述預設環境參數可以包括光圈、曝光值、光線角度,預設環境參數可以設定一閾值或區間值,以篩選出具有不同方位角度且畫質較清晰的圖像。篩選模組50根據拍攝的方位角度及預設環境參數從第一提取模組40提取出的200張圖像中篩選出符合生成全景影像的第二圖像。The screening module 50 is configured to filter, from the extracted first image, a second image having different azimuth angles and conforming to preset environmental parameters. For example, the preset environment parameter may include an aperture, an exposure value, and a ray angle, and the preset environment parameter may set a threshold or an interval value to filter out images having different azimuth angles and clearer image quality. The screening module 50 selects a second image that is consistent with the generated panoramic image from the 200 images extracted from the first extraction module 40 according to the captured azimuth angle and the preset environmental parameters.

在一實施方式中,全景影像優選由不同方位角度且畫質較清晰的第二圖像拼接而成。In an embodiment, the panoramic image is preferably spliced by a second image with different azimuth angles and clearer image quality.

第二提取模組60用於根據所述每一第二圖像的環境參數與圖像參數從中提取出各方位角度的優選圖像。所述每一方位角度的優選圖像可以是相鄰方位角度中環境參數與圖像參數差異最小的一第二圖像。舉例而言,經過篩選模組50的篩選,方位角度A具有3張圖像,方位角度B具有5張圖像,方位角度C具有4張圖像,方位角度D具有7張圖像。若拼接所述全景影像需要一張方位角度A圖像、一張方位角度B圖像、一張方位角度C圖像及一張方位角度D圖像。此時,第二提取模組60根據拼接的相鄰圖像的環境參數與圖像參數來作為參照從多張方位角度A圖像、方位角度B圖像、方位角度C圖像、方位角度D圖像中進行挑選,從而避免相鄰的拼接圖像的參數相差較大,降低了視覺效果。舉例而言,第二提取模組60根據與方位角度A相鄰的拼接圖像的參數來從3張方位角度A圖像中提取出一張與方位角度A相鄰的拼接圖像的參數差異最小的圖像作為方位角度A的優選圖像。在本發明的其他實施方式中,第二提取模組60也可以從3張方位角度A圖像中提取出環境參數和/或圖像參數最佳的一張方位角度A圖像作為優選圖像。The second extraction module 60 is configured to extract a preferred image of the respective angles from the environment parameters and the image parameters of each of the second images. The preferred image of each azimuth angle may be a second image in which the difference between the environmental parameter and the image parameter in the adjacent azimuth angle is the smallest. For example, after screening by the screening module 50, the azimuth angle A has three images, the azimuth angle B has five images, the azimuth angle C has four images, and the azimuth angle D has seven images. If the panoramic image is spliced, an azimuth angle A image, an azimuth angle B image, an azimuth angle C image, and an azimuth angle D image are required. At this time, the second extraction module 60 takes as reference from the plurality of azimuth angle A images, the azimuth angle B image, the azimuth angle C image, and the azimuth angle D according to the environmental parameters and image parameters of the spliced adjacent images. The images are selected to avoid large differences in parameters of adjacent stitched images and reduce visual effects. For example, the second extraction module 60 extracts a parameter difference of the stitched image adjacent to the azimuth angle A from the three azimuth angle A images according to the parameters of the stitched image adjacent to the azimuth angle A. The smallest image is taken as the preferred image of the azimuth angle A. In other embodiments of the present invention, the second extraction module 60 may also extract an azimuth angle A image with the best environmental parameters and/or image parameters from the three azimuth angle A images as a preferred image. .

拼接生成模組70用於調整所述每一優選圖像的圖像參數以進行拼接生成一全景影像。具體地,拼接生成模組70可以調整第二提取模組60提取出的每一優選圖像的顏色、明亮度、銳利度、焦距等。拼接生成模組70再對調整後的圖像進行拼接以生成所述全景影像。The splicing generation module 70 is configured to adjust image parameters of each of the preferred images to perform splicing to generate a panoramic image. Specifically, the splicing generation module 70 can adjust the color, brightness, sharpness, focal length, and the like of each preferred image extracted by the second extraction module 60. The stitching generation module 70 then stitches the adjusted image to generate the panoramic image.

請同時參閱圖3-4,在一實施方式中,拼接生成模組70包括調整單元702及拼接單元704。由於每一優選照片在拍攝時可能焦距不同,此時若直接進行拼接將導致視覺效果不佳。調整單元702用於將所述每一優選圖像進行中心點校正以調整所述每一優選圖像的焦距,使得焦距基本一致,進行焦距調整時,圖像比例和角度亦會經過適當調整。調整單元702同時調整所述每一優選圖像的圖像參數。拼接單元704用於選擇焦距最接近校正後的中心點的優選圖像作為起始拼接圖像並以此圖像基礎向四周進行拼接,以生成所述全景影像。Referring to FIG. 3-4 at the same time, in an embodiment, the splicing generation module 70 includes an adjustment unit 702 and a splicing unit 704. Since each preferred photo may have a different focal length when shooting, if the stitching is directly performed, the visual effect will be poor. The adjusting unit 702 is configured to perform center point correction on each of the preferred images to adjust the focal length of each of the preferred images such that the focal lengths are substantially the same, and the image ratio and angle are also appropriately adjusted when the focus adjustment is performed. The adjustment unit 702 simultaneously adjusts the image parameters of each of the preferred images. The splicing unit 704 is configured to select a preferred image whose focal length is closest to the corrected center point as a starting stitching image and splicing to the periphery based on the image base to generate the panoramic image.

在一實施方式中,接收模組20接收的基礎檢索資訊也可以為一使用者移動裝置(手機、平板電腦)中相冊的任意一圖像A。全景影像生成裝置100還包括判斷模組80及標記模組90。所述全景影像可以是180度全景影像、270度全景影像、360度全景影像等多種類型。In an embodiment, the basic search information received by the receiving module 20 may also be any image A of an album in a user mobile device (mobile phone, tablet). The panoramic image generating device 100 further includes a determining module 80 and a marking module 90. The panoramic image may be of a plurality of types such as a 180-degree panoramic image, a 270-degree panoramic image, and a 360-degree panoramic image.

判斷模組80用於判斷第二提取模組60提取的優選圖像是否足夠生成所述任意一角度類型的全景影像。若判斷模組80判斷所述提取的優選圖像能夠生成一種或者多種全景影像時,標記模組90在所述相冊的圖像A上添加一可生成的最大角度類型的標記資訊,以提示使用者此圖像A的拍攝地點有全景影像可供觀看。舉例而言,若判斷模組80判斷所述提取的優選圖像能夠生成180度、270度、360度三種全景影像。標記模組90在所述相冊的圖像A上添加一可生成360度全景影像的標記資訊。若判斷模組80判斷所述提取的優選圖像能夠生成180度、270度二種全景影像。標記模組90在所述相冊的圖像A上添加一可生成270度全景影像的標記資訊。使用者可以點擊相冊中的圖片A或圖片A上的標記資訊,後臺即進行圖像拼接,以生成圖像A對應的全景影像供觀看。The determining module 80 is configured to determine whether the preferred image extracted by the second extraction module 60 is sufficient to generate the panoramic image of the arbitrary angle type. If the determining module 80 determines that the extracted preferred image is capable of generating one or more panoramic images, the marking module 90 adds a markable information of a maximum angle type that can be generated on the image A of the album to prompt use. The image of this image A has a panoramic image for viewing. For example, if the determination module 80 determines that the extracted preferred image is capable of generating three panoramic images of 180 degrees, 270 degrees, and 360 degrees. The marking module 90 adds a marker information that can generate a 360-degree panoramic image on the image A of the album. If the determining module 80 determines that the extracted preferred image is capable of generating two panoramic images of 180 degrees and 270 degrees. The marking module 90 adds a marker information that can generate a 270-degree panoramic image on the image A of the album. The user can click on the picture A in the album or the mark information on the picture A, and the image is stitched in the background to generate the panoramic image corresponding to the image A for viewing.

當判斷模組80判斷所述提取的優選圖像不能夠生成所述任意一角度類型的全景影像時,所述標記模組90不對所述相冊的圖像A進行標記處理或所述標記模組90在所述相冊的圖像A上添加一不可生成全景影像的標記資訊。When the determining module 80 determines that the extracted preferred image is not capable of generating the panoramic image of any angle type, the marking module 90 does not perform marking processing or the marking module on the image A of the album. 90 adds a markup information that cannot generate a panoramic image on the image A of the album.

在一實施方式中,由於全景影像是由不同方位角度的優選圖像拼接而成,判斷模組80判斷提取的優選圖像是否足夠生成所述任意一角度類型的全景影像的判斷依據可以是判斷生成一角度類型的全景影像是否缺失某一方位角度圖像,若判斷為缺失,則認為不能生成該角度類型的全景影像。In an embodiment, since the panoramic image is formed by splicing the preferred images of different azimuth angles, the determining module 80 determines whether the extracted preferred image is sufficient to generate the panoramic image of the arbitrary angle type. A panoramic image of an angle type is generated to delete an azimuth angle image. If it is judged to be missing, it is considered that the panoramic image of the angle type cannot be generated.

在一實施方式中,若拼接生成模組70在生成一角度類型的全景影像存在遺缺且遺缺內容較小時,可以通過內容感知技術進行填補。In an embodiment, if the splicing generation module 70 generates a panoramic image of an angle type and the missing content is small, the content can be filled by the content sensing technology.

圖5為本發明一實施方式中全景影像生成方法的流程圖。本方法可以使用在圖2或圖3所示的全景影像生成裝置100中。FIG. 5 is a flowchart of a method for generating a panoramic image according to an embodiment of the present invention. The method can be used in the panoramic image generating device 100 shown in FIG. 2 or 3.

步驟S500,接收模組20接收一基礎檢索資訊。In step S500, the receiving module 20 receives a basic search information.

步驟S502,解析模組30根據所述基礎檢索資訊解析出所需檢索的圖像資訊。In step S502, the analysis module 30 parses the image information that needs to be retrieved according to the basic search information.

步驟S504,第一提取模組40從圖像資料庫10中提取出符合所述圖像資訊的第一圖像。In step S504, the first extraction module 40 extracts a first image that conforms to the image information from the image database 10.

步驟S506,篩選模組50從所述提取的第一圖像中篩選出具有不同方位角度且符合預設環境參數的第二圖像。Step S506, the screening module 50 filters out the second image having different azimuth angles and conforming to the preset environmental parameters from the extracted first image.

步驟S508,第二提取模組60根據所述每一第二圖像的環境參數與圖像參數從中提取出各方位角度的優選圖像。Step S508, the second extraction module 60 extracts a preferred image of the respective angles from the environment parameters and the image parameters of each of the second images.

步驟S510,拼接生成模組70調整所述每一優選圖像的圖像參數以進行拼接生成一全景影像。In step S510, the splicing generation module 70 adjusts the image parameters of each of the preferred images to perform splicing to generate a panoramic image.

在一實施方式中,拼接生成模組70包括調整單元702及拼接單元704。由於每一優選照片在拍攝時可能焦距不同,此時若直接進行拼接將導致視覺效果不佳。調整單元702將所述每一優選圖像進行中心點校正以調整所述每一優選圖像的焦距,使得焦距基本一致,進行焦距調整時,圖像比例和角度亦會經過適當調整。調整單元702同時調整所述每一優選圖像的圖像參數。拼接單元704選擇焦距最接近校正後的中心點的優選圖像作為起始拼接圖像並以此圖像基礎向四周進行拼接,以生成所述全景影像。In an embodiment, the splicing generation module 70 includes an adjustment unit 702 and a splicing unit 704. Since each preferred photo may have a different focal length when shooting, if the stitching is directly performed, the visual effect will be poor. The adjusting unit 702 performs center point correction on each of the preferred images to adjust the focal length of each of the preferred images such that the focal lengths are substantially the same, and the image ratio and angle are also appropriately adjusted when the focus adjustment is performed. The adjustment unit 702 simultaneously adjusts the image parameters of each of the preferred images. The splicing unit 704 selects a preferred image whose focal length is closest to the corrected center point as a starting spliced image and splicing to the periphery based on the image base to generate the panoramic image.

圖6為本發明另一實施方式中全景影像生成方法的流程圖。圖6與圖5所示的流程步驟基本相同,不同之處在於:FIG. 6 is a flowchart of a method for generating a panoramic image according to another embodiment of the present invention. The process steps shown in Figure 6 and Figure 5 are basically the same, except that:

S500,接收模組20接收一基礎檢索資訊,所述基礎檢索資訊為一相冊中的圖像。圖6所揭示的方法還包括步驟S512、步驟S514及步驟S516。S500. The receiving module 20 receives a basic search information, where the basic search information is an image in an album. The method disclosed in FIG. 6 further includes step S512, step S514, and step S516.

步驟S512,判斷模組80判斷所述提取的優選圖像是否足夠生成任意一預設角度類型的全景影像(180度、270度、360度)。In step S512, the determining module 80 determines whether the extracted preferred image is sufficient to generate a panoramic image of any preset angle type (180 degrees, 270 degrees, 360 degrees).

步驟S514,若判斷模組80判斷能夠生成一種或多種預設角度全景影像,標記模組90在所述相冊的圖像上添加一可生成的最大角度類型的標記資訊。Step S514, if the determining module 80 determines that one or more preset angle panoramic images can be generated, the marking module 90 adds a markable information of a maximum angle type that can be generated on the image of the album.

步驟S516,若判斷模組80判斷不能夠生成所述任意一種預設角度全景影像,標記模組90不對所述相冊的圖像進行標記處理。In step S516, if the determining module 80 determines that the preset one-angle panoramic image cannot be generated, the marking module 90 does not perform marking processing on the image of the album.

在一實施方式中,若判斷模組80判斷不能夠生成所述任意一種預設角度全景影像,所述標記模組90也可以在所述相冊的圖像上添加一不可生成全景影像的標記資訊。In an embodiment, if the determining module 80 determines that the preset one-angle panoramic image cannot be generated, the marking module 90 may also add a marking information that cannot generate the panoramic image on the image of the album. .

上述全景影像生成裝置及方法,可以根據接收到的基礎檢索資訊(語音/文字/圖像)來篩選圖像並將圖像進行拼接生成全景影像供使用者觀看,提示用戶的觀看體驗。The panoramic image generating device and method can filter images according to the received basic search information (speech/text/image) and splicing the images to generate a panoramic image for the user to view, prompting the user's viewing experience.

綜上所述,本發明符合發明專利要件,爰依法提出專利申請。惟,以上所述者僅為本發明之較佳實施方式,本發明之範圍並不以上述實施方式為限,舉凡熟悉本案技藝之人士爰依本發明之精神所作之等效修飾或變化,皆應涵蓋於以下申請專利範圍內。In summary, the present invention complies with the requirements of the invention patent and submits a patent application according to law. However, the above description is only a preferred embodiment of the present invention, and the scope of the present invention is not limited to the above-described embodiments, and equivalent modifications or variations made by those skilled in the art in accordance with the spirit of the present invention are It should be covered by the following patent application.

10‧‧‧圖像資料庫 10‧‧‧ Image database

20‧‧‧接收模組 20‧‧‧ receiving module

30‧‧‧解析模組 30‧‧‧Analytical Module

40‧‧‧第一提取模組 40‧‧‧First extraction module

50‧‧‧篩選模組 50‧‧‧Screening module

60‧‧‧第二提取模組 60‧‧‧Second extraction module

70‧‧‧拼接生成模組 70‧‧‧Splicing generation module

80‧‧‧判斷模組 80‧‧‧Judgement module

90‧‧‧標記模組 90‧‧‧Marking module

100‧‧‧全景影像生成裝置 100‧‧‧ panoramic image generating device

702‧‧‧調整單元 702‧‧‧Adjustment unit

704‧‧‧拼接單元 704‧‧‧Splicing unit

圖1是本發明一實施方式的全景影像生成裝置的運用環境圖。1 is a diagram showing an operational environment of a panoramic image generating device according to an embodiment of the present invention.

圖2是本發明一實施方式的全景影像生成裝置的功能模組圖。2 is a functional block diagram of a panoramic image generating device according to an embodiment of the present invention.

圖3是本發明另一實施方式的全景影像生成裝置的功能模組圖。3 is a functional block diagram of a panoramic image generating device according to another embodiment of the present invention.

圖4是本發明一實施方式的多張優選圖像進行中心點校正的示意圖。4 is a schematic diagram of center point correction of a plurality of preferred images according to an embodiment of the present invention.

圖5是本發明一實施方式的全景影像生成方法的步驟流程圖。FIG. 5 is a flow chart showing the steps of a method for generating a panoramic image according to an embodiment of the present invention.

圖6是本發明另一實施方式的全景影像生成方法的步驟流程圖。6 is a flow chart showing the steps of a method for generating a panoramic image according to another embodiment of the present invention.

no

no

Claims (12)

一種全景影像生成方法,所述全景影像生成方法包括: 接收一基礎檢索資訊; 根據所述基礎檢索資訊解析出所需檢索的圖像資訊; 從圖像資料庫中提取出符合所述圖像資訊的第一圖像; 從所述提取的第一圖像中篩選出具有不同方位角度且符合預設環境參數的第二圖像; 根據所述每一第二圖像的環境參數與圖像參數從中提取出各方位角度的優選圖像;及 調整所述每一優選圖像的圖像參數以進行拼接生成一全景影像。A method for generating a panoramic image, the method for generating a panoramic image, comprising: receiving a basic search information; parsing image information to be retrieved according to the basic search information; and extracting the image information from the image database a first image; a second image having different azimuth angles and conforming to preset environmental parameters is filtered out from the extracted first image; and environmental parameters and image parameters according to each of the second images Extracting a preferred image of the angles of the respective parties; and adjusting image parameters of each of the preferred images to perform splicing to generate a panoramic image. 如申請專利範圍第1項所述之全景影像生成方法,其中所述調整所述每一優選圖像的圖像參數以進行拼接生成所述全景影像的步驟包括: 將所述每一優選圖像進行中心點校正,以調整所述每一優選圖像的焦距; 調整所述每一優選圖像的圖像參數; 選擇焦距最接近校正後的中心點的優選圖像作為起始拼接圖像向四周進行拼接,以生成所述全景影像。The method for generating a panoramic image according to claim 1, wherein the adjusting the image parameters of each of the preferred images to perform splicing to generate the panoramic image comprises:: each of the preferred images Performing a center point correction to adjust a focal length of each of the preferred images; adjusting image parameters of each of the preferred images; selecting a preferred image having a focal length closest to the corrected center point as a starting stitching image orientation Stitching is performed around to generate the panoramic image. 如申請專利範圍第1項所述之全景影像生成方法,其中所述基礎檢索資訊包括圖像、語音資訊、文字資訊中的一種或多種。The method for generating a panoramic image according to claim 1, wherein the basic search information includes one or more of image, voice information, and text information. 如申請專利範圍第1項所述之全景影像生成方法,其中所述每一方位角度的優選圖像是相鄰方位角度中環境參數與圖像參數差異最小的第二圖像。The method for generating a panoramic image according to claim 1, wherein the preferred image of each azimuth angle is a second image in which the difference between the environmental parameter and the image parameter in the adjacent azimuth angle is the smallest. 如申請專利範圍第1項所述之全景影像生成方法,其中所述基礎檢索資訊為一相冊中的圖像,所述全景影像包括多種預設角度類型的全景影像,所述全景影像生成方法還包括: 判斷所述提取的優選圖像是否足夠生成所述任意一預設角度類型的全景影像;及 若能夠生成所述全景影像,在所述相冊的圖像上添加一可生成的最大角度類型的標記資訊。The method for generating a panoramic image according to claim 1, wherein the basic search information is an image in an album, and the panoramic image includes a plurality of preset angle types of panoramic images, and the panoramic image generating method further The method includes: determining whether the extracted preferred image is sufficient to generate the panoramic image of any one of the preset angle types; and if the panoramic image can be generated, adding a maximum angle type that can be generated on the image of the album Tag information. 如申請專利範圍第5項所述之全景影像生成方法,其中所述全景影像生成方法還包括: 若不能夠生成所述全景影像,不對所述相冊的圖像進行標記處理或在所述相冊的圖像上添加一不可生成全景影像的標記資訊。The method for generating a panoramic image according to claim 5, wherein the method for generating a panoramic image further comprises: if the panoramic image cannot be generated, not marking the image of the album or in the album Add a marker information on the image that cannot generate a panoramic image. 一種全景影像生成裝置,包括一圖像資料庫,所述圖像資料庫存儲有多張圖像,其改良在於,所述全景影像生成裝置還包括: 接收模組,用於接收一基礎檢索資訊; 解析模組,用於根據所述基礎檢索資訊解析出所需檢索的圖像資訊; 第一提取模組,用於從所述圖像資料庫中提取出符合所述圖像資訊的第一圖像; 篩選模組,用於從所述提取的第一圖像中篩選出具有不同方位角度且符合預設環境參數的第二圖像; 第二提取模組,用於根據所述每一第二圖像的環境參數與圖像參數從中提取出各方位角度的優選圖像;及 拼接生成模組,用於調整所述每一優選圖像的圖像參數以進行拼接生成一全景影像。A panoramic image generating device, comprising an image database, wherein the image data library stores a plurality of images, wherein the panoramic image generating device further comprises: a receiving module, configured to receive a basic search information The parsing module is configured to parse the image information to be retrieved according to the basic search information; and the first extracting module is configured to extract, from the image database, the first image that meets the image information a screening module, configured to filter, from the extracted first image, a second image that has different azimuth angles and meets preset environmental parameters; and a second extraction module, configured to The environment parameter and the image parameter of the second image are used to extract a preferred image of the angles of the respective parties; and the splicing generation module is configured to adjust the image parameters of each of the preferred images to perform splicing to generate a panoramic image. 如申請專利範圍第7項所述之全景影像生成裝置,其中所述拼接生成模組包括調整單元及拼接單元,所述調整單元用於將所述每一優選圖像進行中心點校正以調整所述每一優選圖像的焦距,並同時調整所述每一優選圖像的圖像參數;所述拼接單元用於選擇焦距最接近校正後的中心點的優選圖像作為起始拼接圖像向四周進行拼接,以生成所述全景影像。The panoramic image generating device of claim 7, wherein the splicing generating module comprises an adjusting unit and a splicing unit, wherein the adjusting unit is configured to perform center point correction on each of the preferred images to adjust the Describe the focal length of each preferred image, and simultaneously adjust the image parameters of each of the preferred images; the splicing unit is configured to select a preferred image whose focal length is closest to the corrected center point as the starting stitching image direction Stitching is performed around to generate the panoramic image. 如申請專利範圍第7項所述之全景影像生成裝置,其中所述基礎檢索資訊包括圖像、語音資訊、文字資訊中的一種或多種。The panoramic image generating apparatus according to claim 7, wherein the basic search information includes one or more of an image, a voice information, and a text information. 如申請專利範圍第7項所述之全景影像生成裝置,其中所述每一方位角度的優選圖像是相鄰方位角度中環境參數與圖像參數差異最小的第二圖像。The panoramic image generating device of claim 7, wherein the preferred image of each azimuth angle is a second image in which an environmental parameter and an image parameter have the smallest difference in adjacent azimuth angles. 如申請專利範圍第7項所述之全景影像生成裝置,其中所述全景影像生成裝置還包括判斷模組及標記模組,所述基礎檢索資訊為一相冊中的圖像,所述全景影像包括多種預設角度類型的全景影像,所述判斷模組用於判斷所述提取的優選圖像是否足夠生成所述任意一預設角度類型的全景影像,所述標記模組用於在判斷所述提取的優選圖像能夠生成所述全景影像時,在所述相冊的圖像上添加一可生成的最大角度類型的標記資訊。The panoramic image generating device of claim 7, wherein the panoramic image generating device further comprises a determining module and a marking module, wherein the basic search information is an image in an album, and the panoramic image includes a plurality of preset angle type panoramic images, the determining module is configured to determine whether the extracted preferred image is sufficient to generate the panoramic image of any one of the preset angle types, and the marking module is configured to determine the When the extracted preferred image is capable of generating the panoramic image, a mark information of a maximum angle type that can be generated is added to the image of the album. 如申請專利範圍第11項所述之全景影像生成裝置,其中當所述判斷模組用於判斷所述提取的優選圖像不能夠生成所述任意一角度類型的全景影像時,所述標記模組不對所述相冊的圖像進行標記處理或所述標記模組在所述相冊的圖像上添加一不可生成全景影像的標記資訊。The panoramic image generating device of claim 11, wherein the marking module is used when the determining module is configured to determine that the extracted preferred image is not capable of generating the panoramic image of any angle type. The group does not mark the image of the album or the tag module adds a markup information on the image of the album that cannot generate a panoramic image.
TW106129172A 2017-08-28 2017-08-28 Device and method for generating panorama image TWI636424B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW106129172A TWI636424B (en) 2017-08-28 2017-08-28 Device and method for generating panorama image

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW106129172A TWI636424B (en) 2017-08-28 2017-08-28 Device and method for generating panorama image

Publications (2)

Publication Number Publication Date
TWI636424B TWI636424B (en) 2018-09-21
TW201913562A true TW201913562A (en) 2019-04-01

Family

ID=64453168

Family Applications (1)

Application Number Title Priority Date Filing Date
TW106129172A TWI636424B (en) 2017-08-28 2017-08-28 Device and method for generating panorama image

Country Status (1)

Country Link
TW (1) TWI636424B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111240175A (en) * 2020-03-13 2020-06-05 联想(北京)有限公司 Projection control method, device and system

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105701767A (en) * 2014-11-28 2016-06-22 富士通株式会社 Panoramic image constructing device and method

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111240175A (en) * 2020-03-13 2020-06-05 联想(北京)有限公司 Projection control method, device and system

Also Published As

Publication number Publication date
TWI636424B (en) 2018-09-21

Similar Documents

Publication Publication Date Title
CN101910936B (en) Guided photography based on image capturing device rendered user recommendations
US9911002B2 (en) Method of modifying image including photographing restricted element, and device and system for performing the method
US8072503B2 (en) Methods, apparatuses, systems, and computer program products for real-time high dynamic range imaging
TWI673682B (en) Method of modifying image including photographing restricted element, and device and system for performing the method
CN105981368A (en) Photo composition and position guidance in an imaging device
US9756260B1 (en) Synthetic camera lenses
KR101653041B1 (en) Method and apparatus for recommending photo composition
KR20090087670A (en) Method and system for extracting the photographing information
WO2016197734A1 (en) Image capturing method, terminal and server
US10958837B2 (en) Systems and methods for determining preferences for capture settings of an image capturing device
CN108028881A (en) Camera auxiliary system, device and method and camera shooting terminal
CN110581950B (en) Camera, system and method for selecting camera settings
CN116195263A (en) Recommending locations and content aware filters for digital photos
CN113824883A (en) Information processing method and electronic equipment
US10860166B2 (en) Electronic apparatus and image processing method for generating a depth adjusted image file
CN110019886A (en) Full-view image generating means and method
US20150341541A1 (en) Methods and systems of remote acquisition of digital images or models
TWI636424B (en) Device and method for generating panorama image
Taheri et al. Focusing and depth of field in photography: application in dermatology practice
DE102018208056A1 (en) Service that implements a workflow based on image analysis
JP4870503B2 (en) Camera and blog management system
JP2006350546A (en) Information processor, image classification method and information processing system
CN112565586A (en) Automatic focusing method and device
JP2004274735A (en) Imaging apparatus and image processing apparatus
US20230216979A1 (en) Information processing device, information processing method, and program