TWI736060B - High-resolution video image processing method, device, and electronic device - Google Patents
High-resolution video image processing method, device, and electronic device Download PDFInfo
- Publication number
- TWI736060B TWI736060B TW108145567A TW108145567A TWI736060B TW I736060 B TWI736060 B TW I736060B TW 108145567 A TW108145567 A TW 108145567A TW 108145567 A TW108145567 A TW 108145567A TW I736060 B TWI736060 B TW I736060B
- Authority
- TW
- Taiwan
- Prior art keywords
- roi
- video image
- key
- image processing
- area
- Prior art date
Links
Images
Landscapes
- Image Processing (AREA)
Abstract
Description
本發明涉及視頻影像處理技術領域,尤其涉及一種高解析度視頻影像處理方法、裝置及電子設備。 The present invention relates to the technical field of video image processing, in particular to a high-resolution video image processing method, device and electronic equipment.
隨著圖像傳感技術與編解碼技術之進步,超高清視頻圖像(4K/8K)之應用日漸普及。超高清圖像細節將極大地提高諸如工業檢驗、監控等多種應用之效果。與此同時,人工智慧AI技術包括深度學習之進步,使得視頻圖像之智慧自動分析處理(例如分類、缺陷識別等)日漸可靠。但對於8K等超高清視頻圖像,資料量是2K全高清視頻之16倍,於進行即時視頻處理時,需要系統有很高之計算能力。 With the advancement of image sensing technology and codec technology, the application of ultra-high-definition video images (4K/8K) is becoming more and more popular. The ultra-high-definition image details will greatly improve the effects of various applications such as industrial inspection and surveillance. At the same time, artificial intelligence AI technology, including advances in deep learning, makes intelligent automatic analysis and processing of video images (such as classification, defect recognition, etc.) increasingly reliable. However, for ultra-high-definition video images such as 8K, the amount of data is 16 times that of 2K full-HD video. For real-time video processing, the system requires high computing power.
另外,雖於8K電視等大屏系統,可做到8K超高清解碼與顯示。但對於很多可擕式終端,例如手機、平板電腦等,超高清解碼與顯示能力仍然不足。因此,如何將智慧AI處理後之結果與視頻圖像,於超高清解碼與顯示能力受限之設備,例如手機上實現超高清細節呈現是目前面臨之一項重要課題。 In addition, 8K ultra-high-definition decoding and display can be achieved in large-screen systems such as 8K TVs. But for many portable terminals, such as mobile phones, tablet computers, etc., ultra-high-definition decoding and display capabilities are still insufficient. Therefore, how to apply the results and video images processed by smart AI to devices with limited ultra-high-definition decoding and display capabilities, such as mobile phones, is an important issue currently facing.
鑒於上述內容,有必要提供一種高解析度視頻影像處理方法、裝置及電子設備。 In view of the foregoing, it is necessary to provide a high-resolution video image processing method, device, and electronic equipment.
本發明一方面提供一種視頻影像處理方法,所述方法包括:(a)採集具有高解析度之視頻圖像;(b)定義所述視頻圖像中一個或複數個感興趣且需要進一步處理之畫面區域為ROI(Region Of Interest),並識別出需要處理之複數ROI區域;(c)對所述複數ROI區域進行分析及處理,以得到每一ROI區域之關鍵資訊,並根據所述關鍵資訊獲得進一步之關鍵ROI區域;以及(d)將獲得之ROI區域及分析結果發送至相應之顯示終端,並於所述顯示終端之螢幕上進行展示。 One aspect of the present invention provides a video image processing method, the method comprising: (a) collecting a high-resolution video image; (b) defining one or more of the video images that are of interest and require further processing The screen area is ROI (Region Of Interest), and the complex ROI area that needs to be processed is identified; (c) The complex ROI area is analyzed and processed to obtain the key information of each ROI area, and based on the key information Obtain a further key ROI area; and (d) send the obtained ROI area and analysis result to the corresponding display terminal, and display it on the screen of the display terminal.
作為一種優選方案,所述方法於執行步驟(b)之前還包括對所述視頻圖像進行預處理之步驟。 As a preferred solution, the method further includes a step of preprocessing the video image before performing step (b).
作為一種優選方案,所述方法於執行步驟(b)之前還包括判斷計算資源是否充分之步驟,當判斷計算資源充分時,執行步驟(b);當判斷計算資源不充分時,先對高解析度之視頻圖像進行處理,以將其轉換為低解析度之視頻圖像,再執行步驟(b)。 As a preferred solution, the method further includes the step of judging whether the computing resources are sufficient before performing step (b). When it is judged that the computing resources are sufficient, step (b) is executed; when it is judged that the computing resources are insufficient, the high-resolution The high-resolution video image is processed to convert it into a low-resolution video image, and then step (b) is performed.
作為一種優選方案,所述方法還包括:根據獲取到之關鍵ROI區域,對所述關鍵ROI區域進行分析及處理,以得到關鍵ROI區域之關聯資訊,並根據所述關聯資訊獲得進一步之關聯ROI區域。 As a preferred solution, the method further includes: analyzing and processing the key ROI area according to the obtained key ROI area to obtain the relevant information of the key ROI area, and obtaining further relevant ROI according to the relevant information area.
作為一種優選方案,所述ROI為全景畫面之一個子畫面、全景畫面本身或降低解析度之全景畫面。 As a preferred solution, the ROI is a sub-picture of the panoramic picture, the panoramic picture itself, or the panoramic picture with reduced resolution.
本發明另一方面提供一種視頻影像處理裝置,所述裝置包括: 採集模組,用以對高解析度之視頻圖像進行採集;第一處理模組,用以定義視頻圖像中一個或複數個感興趣且需要進一步處理之畫面區域為ROI(Region Of Interest),並識別出需要處理之複數ROI區域;第二處理模組,用以對所述複數ROI區域進行分析及處理,以得到每一ROI區域之關鍵資訊,並根據所述關鍵資訊獲得進一步之關鍵ROI區域;以及分發模組,用以將獲得之ROI區域及分析結果發送至相應之顯示終端,並於所述顯示終端之螢幕上進行展示。 Another aspect of the present invention provides a video image processing device, the device comprising: The acquisition module is used to collect high-resolution video images; the first processing module is used to define one or more regions of the video image that are of interest and require further processing as ROI (Region Of Interest) , And identify the complex ROI area that needs to be processed; the second processing module is used to analyze and process the complex ROI area to obtain the key information of each ROI area, and obtain further key information based on the key information ROI area; and a distribution module for sending the obtained ROI area and analysis results to the corresponding display terminal, and displaying it on the screen of the display terminal.
作為一種優選方案,所述採集模組還用以對所述視頻圖像進行預處理。 As a preferred solution, the acquisition module is also used to preprocess the video image.
作為一種優選方案,所述視頻影像處理裝置還包括判斷模組及降解析度模組,所述判斷模組用以判斷計算資源是否充分,當判斷計算資源不充分時,所述降解析度模組用以將高解析度之視頻圖像進行處理,以將其轉換為低解析度之視頻圖像。 As a preferred solution, the video image processing device further includes a judgment module and a down-resolution module. The judgment module is used to judge whether the computing resources are sufficient. When it is judged that the computing resources are insufficient, the down-resolution module is The group is used to process high-resolution video images to convert them into low-resolution video images.
作為一種優選方案,所述視頻影像處理裝置還包括第三處理模組,用以根據獲取到之關鍵ROI區域,對所述關鍵ROI區域進行分析及處理,以得到關鍵ROI區域之關聯資訊,並根據所述關聯資訊獲得進一步之關聯ROI區域。 As a preferred solution, the video image processing device further includes a third processing module for analyzing and processing the key ROI area according to the acquired key ROI area to obtain related information of the key ROI area, and A further associated ROI area is obtained according to the associated information.
作為一種優選方案,所述ROI為全景畫面之一個子畫面、全景畫面本身或降低解析度之全景畫面。 As a preferred solution, the ROI is a sub-picture of the panoramic picture, the panoramic picture itself, or the panoramic picture with reduced resolution.
本發明另一方面提供一種電子設備,所述電子設備執行如上 述任一項項所述之視頻影像處理方法。 Another aspect of the present invention provides an electronic device that performs the above The video image processing method described in any one of the above items.
本發明之視頻影像處理方法及裝置可直接對高解析度之視頻圖像或低解析度之視頻圖像進行快速識別。另外,所述視頻影像處理方法可按各種配置方法,動態獲取各個階段之基於高解析度圖像之細節ROI與分析經過,亦可獲取低解析度之全景預覽。例如,所述視頻影像處理方法可獲得並於小屏之顯示終端顯示具備8K程度細節之某笑臉之ROI、某哭泣之臉之ROI或有出血部分之ROI,亦可瀏覽全景低解析度圖片裡之所有人臉分佈與分析介紹,並點擊進入人臉細節瀏覽。如此,於不需要8K大顯示幕之情況下,利用本發明之視頻影像處理方法,所述顯示終端亦能夠迅速地於8K全幅圖像中訪問到需要關注之ROI圖片,並快速跟進處理。 The video image processing method and device of the present invention can directly quickly identify high-resolution video images or low-resolution video images. In addition, the video image processing method can dynamically obtain the detailed ROI and analysis process based on the high-resolution image at each stage according to various configuration methods, and can also obtain a low-resolution panoramic preview. For example, the video image processing method can obtain and display the ROI of a smiling face with 8K level of detail on the display terminal of the small screen, the ROI of a crying face, or the ROI of a bleeding part, or browse the panoramic low-resolution pictures Introduction to the distribution and analysis of all human faces, and click to enter the face details browsing. In this way, when an 8K large display screen is not required, using the video image processing method of the present invention, the display terminal can also quickly access the ROI picture that needs attention in the 8K full-frame image, and quickly follow up processing.
100:視頻影像處理裝置 100: Video image processing device
11:採集模組 11: Acquisition module
13:判斷模組 13: Judgment module
15:降解析度模組 15: Down-resolution module
16:第一處理模組 16: The first processing module
17:第二處理模組 17: The second processing module
18:第三處理模組 18: The third processing module
19:分發模組 19: Distribution module
200:電子設備 200: electronic equipment
201:記憶體 201: Memory
202:處理器 202: processor
203:電腦程式 203: Computer Program
205:攝影模組 205: Photography module
300:視頻影像處理系統 300: Video image processing system
301:顯示終端 301: display terminal
圖1為本發明較佳實施方式中視頻影像處理方法之流程圖。 FIG. 1 is a flowchart of a video image processing method in a preferred embodiment of the present invention.
圖2為圖2所示視頻影像處理方法之應用場景示意圖。 FIG. 2 is a schematic diagram of application scenarios of the video image processing method shown in FIG. 2.
圖3為本發明較佳實施方式中視頻影像處理裝置之功能框圖。 Fig. 3 is a functional block diagram of a video image processing device in a preferred embodiment of the present invention.
圖4為本發明較佳實施方式中視頻影像處理系統之功能框圖。 4 is a functional block diagram of the video image processing system in the preferred embodiment of the present invention.
下面將結合本發明實施例中之附圖,對本發明實施例中之技術方案進行清楚、完整地描述,顯然,所描述之實施例僅僅是本發明一部分實施例,而不是全部之實施例。基於本發明中之實施例,所屬領域具有通常知識者於沒有做出創造性勞動前提下所獲得之所有其他實施例,均屬於本發明保護之範圍。 The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those with ordinary knowledge in the field without creative work shall fall within the protection scope of the present invention.
需要說明的是,當一個元件被稱為“電連接”另一個元件,它可直接於另一個元件上或者亦可存在居中之元件。當一個元件被認為是“電連接`”另一個元件,它可是接觸連接,例如,可是導線連接之方式,亦可是非接觸式連接,例如,可是非接觸式耦合之方式。 It should be noted that when an element is referred to as being "electrically connected" to another element, it can be directly connected to the other element or a central element may also exist. When an element is considered to be "electrically connected" to another element, it can be a contact connection, for example, a wire connection, or a non-contact connection, for example, a non-contact coupling.
除非另有定義,本文所使用之所有之技術與科學術語與屬於所屬領域具有通常知識者通常理解之含義相同。本文中於本發明之說明書中所使用之術語僅是為描述具體之實施例之目不是旨在於限制本發明。 Unless otherwise defined, all technical and scientific terms used in this article have the same meanings commonly understood by those with ordinary knowledge in the field. The terms used in the description of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.
下面結合附圖,對本發明之一些實施方式作詳細說明。於不衝突之情況下,下述之實施例及實施例中之特徵可相互組合。 Hereinafter, some embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the case of no conflict, the following embodiments and the features in the embodiments can be combined with each other.
請參閱圖1,圖1是本發明較佳實施例中視頻影像處理方法之流程圖。該方法用以對具有高解析度之視頻圖像進行有效處理。可理解,根據不同之需求,所述流程圖中步驟之順序可改變,某些步驟可省略。 Please refer to FIG. 1. FIG. 1 is a flowchart of a video image processing method in a preferred embodiment of the present invention. This method is used to effectively process high-resolution video images. It can be understood that, according to different requirements, the order of the steps in the flowchart can be changed, and some steps can be omitted.
步驟S01,即時採集具有高解析度之視頻圖像,並對所述視頻圖像進行預處理。 In step S01, a high-resolution video image is collected in real time, and the video image is preprocessed.
可理解,於本實施例中,可藉由一攝影模組對拍攝視野內之畫面進行圖像傳感,以得到數位化之原始圖元資料構成之畫面。可理解,所述畫面通常是大資料量之4K/8K超高清畫面資料。亦就是說,於本實施例中,可對超高清,例如4K或8K之視頻圖像進行即時採集。 It can be understood that, in this embodiment, a camera module can be used to image the frame in the shooting field of view to obtain a frame composed of digitized original image element data. It can be understood that the picture is usually 4K/8K ultra-high-definition picture data with a large amount of data. In other words, in this embodiment, ultra-high-definition, such as 4K or 8K video images can be captured in real time.
可理解,於本實施例中,可定期從攝影模組取得一幀由原始圖元資料構成之超高清畫面,並根據應用需要將畫面之圖元資料格式做相應之轉換與預先處理。例如,根據配置,每秒採集之幀數可能是25、30、60、120等。圖元之格式可能是UYUV444、YUV444、YUV422、YUV420 等。另外,如有需要,可根據需要對畫面做降噪或邊緣銳化等預處理操作。 It can be understood that, in this embodiment, a frame of ultra-high-definition picture composed of original primitive data can be periodically obtained from the camera module, and the picture primitive data format of the picture can be converted and pre-processed according to the needs of the application. For example, depending on the configuration, the number of frames collected per second may be 25, 30, 60, 120, etc. The format of the primitive may be UYUV444, YUV444, YUV422, YUV420 Wait. In addition, if necessary, pre-processing operations such as noise reduction or edge sharpening can be performed on the picture as needed.
步驟S02,判斷計算資源是否充分。若否,則執行步驟S03。若是,則執行步驟S04。 In step S02, it is judged whether the computing resources are sufficient. If not, go to step S03. If yes, go to step S04.
步驟S03,將高解析度之視頻圖像進行處理,以將其轉換為低解析度之視頻圖像。 Step S03, processing the high-resolution video image to convert it into a low-resolution video image.
例如將8K圖像降低到2K圖像。 For example, the 8K image is reduced to a 2K image.
步驟S04,定義視頻圖像中一個或複數個感興趣且需要進一步處理之畫面區域為ROI(Region Of Interest),並於粗略模式畫面下,識別出需要處理之複數ROI區域。 Step S04: Define one or more regions of interest in the video image that need further processing as ROI (Region Of Interest), and identify the multiple ROI regions that need to be processed in the rough mode.
可理解,ROI可是全景畫面之一個子畫面。此外,全景畫面本身或降低解析度之全景畫面亦可視為一類特別之ROI畫面。 Understandably, the ROI is a sub-picture of the panoramic picture. In addition, the panoramic image itself or the reduced resolution panoramic image can also be regarded as a special kind of ROI image.
可理解,於本實施例中,可定義人臉區域為ROI區域,並利用快速人臉識別演算法迅速標記出視頻圖像中之複數人臉區域。 It can be understood that, in this embodiment, the face area can be defined as an ROI area, and a fast face recognition algorithm is used to quickly mark the plural face areas in the video image.
步驟S05,對所述粗略模式畫面下之複數ROI區域進行分析及處理,以執行精細模式,進而得到每一ROI區域之關鍵資訊,並根據所述關鍵資訊獲得進一步之關鍵ROI區域。 Step S05: Analyze and process the multiple ROI regions in the coarse mode screen to execute the fine mode to obtain key information of each ROI region, and obtain further key ROI regions based on the key information.
例如,針對每個ROI區域(例如每個人臉區域),執行進一步分析處理,進而得到每一ROI區域之關鍵資訊,例如人臉區域之情緒資訊。所述情緒資訊可為歡樂、平靜、哭泣或其他。再根據每一ROI區域之關鍵資訊獲得相應之關鍵ROI區域。例如,根據上述情緒資訊中之哭泣資訊獲得相應之關鍵ROI區域為具有哭泣資訊之人臉區域。 For example, for each ROI area (for example, each face area), further analysis is performed to obtain key information of each ROI area, such as emotional information of the face area. The emotional information can be joy, peace, crying, or others. Then obtain the corresponding key ROI area according to the key information of each ROI area. For example, according to the crying information in the above emotion information, the corresponding key ROI area is obtained as the face area with crying information.
步驟S06,根據獲取到之關鍵ROI區域,對所述關鍵ROI區 域進行分析及處理,以執行關聯模式,進而得到關鍵ROI區域之關聯資訊,並根據所述關聯資訊獲得進一步之關聯ROI區域。 Step S06: According to the acquired key ROI area, The domain is analyzed and processed to execute the correlation mode to obtain the correlation information of the key ROI area, and obtain the further correlation ROI area based on the correlation information.
例如,於步驟S05中,當分析發現哭泣之人臉區域後,進一步推算出與該臉部區域相關之人體輪廓區域,並對該人體輪廓區域進行分析檢測,以判斷是否有出血、跌倒姿態、球衣之號碼等關聯資訊。 For example, in step S05, when the crying face area is found in the analysis, the human contour area related to the face area is further calculated, and the human contour area is analyzed and detected to determine whether there is bleeding, falling posture, Related information such as the number of the jersey.
步驟S07,根據系統組態或一顯示終端之動態需求,將步驟S03、步驟S04、步驟S05與/或步驟S06獲得之ROI圖像區域及分析結果發送至相應之顯示終端,並於所述顯示終端之螢幕上進行展示。 Step S07: According to the system configuration or the dynamic demand of a display terminal, the ROI image area and analysis result obtained in step S03, step S04, step S05 and/or step S06 are sent to the corresponding display terminal and displayed on the Display on the screen of the terminal.
可理解,於其他實施例中,當無需進行關聯區域分析時,可直接執行步驟S07,而省略步驟S06。 It can be understood that, in other embodiments, when the correlation area analysis is not required, step S07 can be directly executed, and step S06 is omitted.
可理解,於本實施例中,所述視頻影像處理方法可直接對高解析度之視頻圖像或低解析度之視頻圖像進行快速識別。另外,所述視頻影像處理方法可按各種配置方法,動態獲取各個階段之基於高解析度圖像之細節ROI與分析經過,亦可獲取低解析度之全景預覽。例如,所述視頻影像處理方法可獲得並於小屏之顯示終端顯示具備8K程度細節之某笑臉之ROI、某哭泣之臉之ROI或有出血部分之ROI,亦可瀏覽全景低解析度圖片裡之所有人臉分佈與分析介紹,並點擊進入人臉細節瀏覽(例如步驟S03)。如此,於不需要8K大顯示幕之情況下,利用本發明之視頻影像處理方法,所述顯示終端亦能夠迅速地於8K全幅圖像中訪問到需要關注之ROI圖片,並快速跟進處理。 It can be understood that, in this embodiment, the video image processing method can directly quickly identify high-resolution video images or low-resolution video images. In addition, the video image processing method can dynamically obtain the detailed ROI and analysis process based on the high-resolution image at each stage according to various configuration methods, and can also obtain a low-resolution panoramic preview. For example, the video image processing method can obtain and display the ROI of a smiling face with 8K level of detail on the display terminal of the small screen, the ROI of a crying face, or the ROI of a bleeding part, or browse the panoramic low-resolution pictures Introduction to the distribution and analysis of all faces, and click to enter the face details browsing (for example, step S03). In this way, when an 8K large display screen is not required, using the video image processing method of the present invention, the display terminal can also quickly access the ROI picture that needs attention in the 8K full-frame image, and quickly follow up processing.
請一併參閱圖2,下面以對人臉區域進行識別及分析為例,對所述視頻影像處理方法進行進一步說明。 Please also refer to FIG. 2. The following takes the recognition and analysis of the face area as an example to further illustrate the video image processing method.
首先,先進行超高清(例如8K)視頻圖像之採集。接著判斷其計算量。當為低計算量時,先對該超高清視頻圖像進行降解析度操作,例如將8K圖像降低到2K圖像。然後執行粗略模式,以識別出相關之複數ROI區域,例如識別出多個人臉區域,包括笑臉、哭泣之臉等。當為高計算量時,則直接跳過降解析度操作,以對該超高清之視頻圖像於粗略模式畫面下,識別出相關之複數ROI區域。 First of all, first carry out the collection of ultra-high-definition (for example, 8K) video images. Then determine the amount of calculation. When the amount of calculation is low, the ultra-high-definition video image is first reduced in resolution, for example, an 8K image is reduced to a 2K image. Then perform the rough mode to identify the relevant complex ROI regions, for example, identify multiple face regions, including smiling faces, crying faces, and so on. When the amount of calculation is high, the down-resolution operation is skipped directly to identify the relevant complex ROI area in the rough mode screen of the ultra-high-definition video image.
接著,對該多個人臉區域進行進一步分析及處理,即執行精細模式,進而獲得每一ROI區域之關鍵資訊,並根據所述關鍵資訊獲得進一步之關鍵ROI區域。例如,對該多個人臉區域進行分析及處理,以獲得每一人臉區域之情緒資訊。所述情緒資訊可為歡樂、平靜、哭泣或其他。再根據每一ROI區域之關鍵資訊獲得相應之關鍵ROI區域。例如,對多個人臉區域之表情或情緒進行分析,並將哭泣之人臉區域作為關鍵之ROI區域。 Then, the multiple face regions are further analyzed and processed, that is, a fine mode is executed to obtain key information of each ROI region, and further key ROI regions are obtained according to the key information. For example, the multiple face regions are analyzed and processed to obtain emotional information of each face region. The emotional information can be joy, peace, crying, or others. Then obtain the corresponding key ROI area according to the key information of each ROI area. For example, analyze the expressions or emotions of multiple face regions, and use the crying face region as the key ROI region.
接著,根據獲取到之關鍵ROI區域,對所述關鍵ROI區域進行分析及處理,以執行關聯模式,進而得到關鍵ROI區域之關聯資訊,並根據所述關聯資訊獲得進一步之關聯ROI區域。例如,當分析發現哭泣之人臉區域後,進一步推算出與該臉部區域相關之關聯人體輪廓區域,並進行分析檢測,以判斷是否有出血、跌倒姿態、球衣之號碼等關聯資訊。 Then, according to the acquired key ROI area, the key ROI area is analyzed and processed to execute the correlation mode, and then the correlation information of the key ROI area is obtained, and further correlation ROI areas are obtained according to the correlation information. For example, when a crying face area is found by analysis, the relevant human contour area related to the face area is further calculated, and analysis is performed to determine whether there is associated information such as bleeding, falling posture, and jersey number.
最後,根據系統組態或顯示終端之動態需求,將各階段獲得之ROI圖像,例如複數ROI區域、關鍵ROI區域與/或關聯ROI區域及分析結果發送至相應之顯示終端,並於所述顯示終端之螢幕上進行展示。 Finally, according to the system configuration or the dynamic requirements of the display terminal, the ROI images obtained at each stage, such as multiple ROI regions, key ROI regions, and/or associated ROI regions, and the analysis results are sent to the corresponding display terminal, and displayed as described Display on the screen of the display terminal.
可理解,請一併參閱圖3,本發明另一實施例還提供一種視頻
影像處理裝置100。所述視頻影像處理裝置100包括採集模組11、判斷模組13、降解析度模組15、第一處理模組16、第二處理模組17、第三處理模組18以及分發模組19。
Understandably, please refer to FIG. 3 together. Another embodiment of the present invention also provides a video
其中,所述採集模組11用以對超高清,例如4K或8K視頻圖像進行即時採集並對所述視頻圖像進行預處理。
Wherein, the
所述判斷模組13用以判斷計算資源是否充分。
The
所述降解析度模組15用以將高解析度之視頻圖像進行處理,以將其轉換為低解析度之視頻圖像。例如將8K圖像降低到2K圖像。
The down-
所述第一處理模組16用以定義視頻圖像中一個或複數個感興趣且需要進一步處理之畫面區域為ROI(Region Of Interest),並於粗略模式畫面下,識別出需要處理之複數ROI區域。
The
可理解,ROI可是全景畫面之一個子畫面。此外,全景畫面本身或降低解析度之全景畫面亦可視為一類特別之ROI畫面。 Understandably, the ROI is a sub-picture of the panoramic picture. In addition, the panoramic image itself or the reduced resolution panoramic image can also be regarded as a special kind of ROI image.
可理解,於本實施例中,可定義人臉區域為ROI區域,並利用快速人臉識別演算法迅速標記出視頻圖像中之複數人臉區域。 It can be understood that, in this embodiment, the face area can be defined as an ROI area, and a fast face recognition algorithm is used to quickly mark the plural face areas in the video image.
所述第二處理模組17用以對所述粗略模式畫面下之複數ROI區域進行分析及處理,以執行精細模式,進而得到每一ROI區域之關鍵資訊,並根據所述關鍵資訊獲得進一步之關鍵ROI區域。例如,針對每個ROI區域(例如每個人臉區域),執行進一步分析處理,進而得到每一ROI區域之關鍵資訊,例如人臉區域之情緒資訊。所述情緒資訊可為歡樂、平靜、哭泣或其他。再根據每一ROI區域之關鍵資訊獲得相應之關鍵ROI區域。例如,根據上述情緒資訊中之哭泣資訊獲得相應之關鍵ROI區域為具有哭
泣資訊之人臉區域。
The
所述第三處理模組18用以根據獲取到之關鍵ROI區域,對所述關鍵ROI區域進行分析及處理,以執行關聯模式,進而得到關鍵ROI區域之關聯資訊,並根據所述關聯資訊獲得進一步之關聯ROI區域。
The
例如,當分析發現哭泣之人臉區域後,進一步推算出與該臉部區域相關之人體輪廓區域,並對該人體輪廓區域進行分析檢測,以判斷是否有出血、跌倒姿態、球衣之號碼等關聯資訊。 For example, when a crying face area is found in the analysis, the human contour area related to the face area is further calculated, and the human contour area is analyzed and detected to determine whether there is any association with bleeding, falling posture, jersey number, etc. News.
所述分發模組19用以根據系統組態或顯示終端之動態需求,將所述第一處理模組16、第二處理模組17及/或第三處理模組18獲得之ROI圖像區域及分析結果發送至相應之顯示終端,並於所述顯示終端之螢幕上進行展示。
The
可理解,請一併參閱圖4,本發明另一實施例還提供一種電子設備200。所述電子設備200包括記憶體201、處理器202以及存儲於所述記憶體201中並可於所述處理器202上運行之電腦程式203。
It is understandable that please refer to FIG. 4 together, another embodiment of the present invention also provides an
所述電子設備200還可包括攝影模組205。所述攝影模組205用以對拍攝視野內之畫面進行圖像傳感,以得到數位化之原始圖元資料構成之畫面。
The
所述電子設備200可為具有超高清視頻影像處理、即時處理器、圖像播放或顯示等功能之裝置,例如8K攝像機、高速伺服器、8K電視之整合系統等。本領域技術人員可理解,所述示意圖僅僅是電子設備200之示例,並不構成對電子設備200之限定,可包括比圖示更多或更少之部件,或者組合某些部件,或者不同之部件。
The
所述處理器202用以執行所述電腦程式203時實現上述視頻影像處理方法實施例中之步驟,例如圖1所示之步驟S01-S07。或者,所述處理器202執行所述電腦程式203時實現上述視頻影像處理裝置100實施例中各模組/單元之功能,例如圖3中之採集模組11、判斷模組13、降解析度模組15、第一處理模組16、第二處理模組17、第三處理模組18以及分發模組19。
The
所述電腦程式203可被分割成一個或多個模組/單元,所述一個或者多個模組/單元被存儲於所述記憶體201中,並由所述處理器202執行,以完成本發明。所述一個或多個模組/單元可是能夠完成特定功能之一系列電腦程式指令段,所述指令段用於描述所述電腦程式203於所述電子設備200中之執行過程。例如,所述電腦程式203可被分割成圖3中之採集模組11、判斷模組13、降解析度模組15、第一處理模組16、第二處理模組17、第三處理模組18以及分發模組19。
The computer program 203 can be divided into one or more modules/units, and the one or more modules/units are stored in the
所述處理器202可是中央處理模組(Central Processing Unit,CPU),還可是其他通用處理器、數位訊號處理器(Digital Signal Processor,DSP)、專用積體電路(Application Specific Integrated Circuit,ASIC)、現成可程式設計閘陣列(Field-Programmable Gate Array,FPGA)或者其他可程式設計邏輯器件、分立門或者電晶體邏輯器件、分立硬體元件等。通用處理器可是微處理器或者所述處理器202亦可是任何常規之處理器等,所述處理器202是所述電子設備200之控制中心,利用各種介面與線路連接整個電子設備200之各個部分。
The
所述記憶體201可用於存儲所述電腦程式203與/或模組/單元.
所述處理器202藉由運行或執行存儲於所述記憶體201內之電腦程式與/或模組/單元,以及調用存儲於記憶體201內之資料,實現所述電子設備200之各種功能。所述記憶體201可主要包括存儲程式區與存儲資料區。其中,存儲程式區可存儲作業系統、至少一個功能所需之應用程式(比如聲音播放功能、圖像播放功能等)等。存儲資料區可存儲根據電子設備200之使用所創建之資料(比如視頻資料、音訊資料、電話本等)等。此外,記憶體201可包括高速隨機存取記憶體,還可包括非易失性記憶體,例如硬碟機、記憶體、插接式硬碟機,智慧存儲卡(Smart Media Card,SMC),安全數位(Secure Digital,SD)卡,快閃記憶體卡(Flash Card)、至少一個磁碟記憶體件、快閃記憶體器件、或其他易失性固態記憶體件。
The
所述電子設備200集成之模組/單元如果以軟體功能模組之形式實現並作為獨立之產品銷售或使用時,可存儲於一個電腦可讀取存儲介質中。基於這樣之理解,本發明實現上述實施例方法中之全部或部分流程,亦可藉由電腦程式來指令相關之硬體來完成,項所述之電腦程式可存儲於一電腦可讀存儲介質中,所述電腦程式於被處理器執行時,可實現上述各個方法實施例之步驟。其中,所述電腦程式包括電腦程式代碼,所述電腦程式代碼可為原始程式碼形式、可執行檔或某些中間形式等。所述電腦可讀介質可包括:能夠攜帶所述電腦程式代碼之任何實體或裝置、記錄介質、U盤、移動硬碟機、磁碟、光碟、電腦記憶體、唯讀記憶體(ROM,Read-Only Memory)、隨機存取記憶體(RAM,Random Access Memory)、電訊號以及軟體分發介質等。需要說明之是,所述電腦可讀介質包含之內容可根據司法管轄區內立法與專利實踐之要求進行適當之增減,例如於某些司法管
轄區,根據立法與專利實踐,電腦可讀介質不包括電載波訊號與電信訊號。
If the integrated module/unit of the
可理解,請再次參閱圖4,本發明另一實施例還提供一種視頻影像處理系統300。所述視頻影像處理系統300包括所述電子設備200及顯示終端301。所述顯示終端301可為手機、平板電腦等超高清解碼與顯示能力受限之終端設備。
Understandably, referring to FIG. 4 again, another embodiment of the present invention also provides a video
於本發明所提供之幾個實施例中,應該理解到,所揭露之電子設備與方法,可藉由其它之方式實現。例如,以上所描述之電子設備實施例僅僅是示意性例如,所述模組之劃分,僅僅為一種邏輯功能劃分,實際實現時可有另外之劃分方式。 In the several embodiments provided by the present invention, it should be understood that the disclosed electronic device and method can be implemented in other ways. For example, the electronic device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division, and there may be other divisions in actual implementation.
另外,於本發明各個實施例中之各功能模組可集成於相同處理模組中,亦可是各個模組單獨物理存於,亦可兩個或兩個以上模組集成於相同模組中。上述集成之模組既可採用硬體之形式實現,亦可採用硬體加軟體功能模組之形式實現。 In addition, the functional modules in the various embodiments of the present invention may be integrated in the same processing module, or each module may be physically stored separately, or two or more modules may be integrated in the same module. The above-mentioned integrated modules can be realized either in the form of hardware or in the form of hardware plus software functional modules.
對於本領域技術人員而言,顯然本發明不限於上述示範性實施例之細節,且於不背離本發明之精神或基本特徵之情況下,能夠以其他之具體形式實現本發明。因此,無論從哪一點來看,均應將實施例看作是示範性且是非限制性本發明之範圍由所附請求項而不是上述說明限定,因此旨於將落於請求項之等同要件之含義與範圍內之所有變化涵括於本發明內。不應將請求項中之任何附圖標記視為限制所涉及之請求項。此外,顯然“包括”一詞不排除其他模組或步驟,單數不排除複數。電子設備請求項中陳述之多個模組或電子設備亦可由同一個模組或電子設備藉由軟體或者硬體來實現。第一,第二等詞語用以表示名稱,而並不表示任何特定之順序。 For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description, and is therefore intended to fall within the equivalent requirements of the claims. All changes within the meaning and scope are included in the present invention. Any reference signs in the request shall not be regarded as the request item involved in the restriction. In addition, it is obvious that the word "include" does not exclude other modules or steps, and the singular does not exclude the plural. Multiple modules or electronic devices stated in the electronic device request can also be implemented by the same module or electronic device by software or hardware. Words such as first and second are used to denote names, but do not denote any specific order.
以上所述,僅為本發明的較佳實施例,並非是對本發明作任何形式上的限定。另外,本領域技術人員還可在本發明精神內做其它變化,當然,這些依據本發明精神所做的變化,都應包含在本發明所要求保護的範圍之內。 The above are only preferred embodiments of the present invention, and are not intended to limit the present invention in any form. In addition, those skilled in the art can also make other changes within the spirit of the present invention. Of course, these changes made according to the spirit of the present invention should all be included in the scope of protection claimed by the present invention.
Claims (9)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW108145567A TWI736060B (en) | 2019-12-12 | 2019-12-12 | High-resolution video image processing method, device, and electronic device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW108145567A TWI736060B (en) | 2019-12-12 | 2019-12-12 | High-resolution video image processing method, device, and electronic device |
Publications (2)
Publication Number | Publication Date |
---|---|
TW202123703A TW202123703A (en) | 2021-06-16 |
TWI736060B true TWI736060B (en) | 2021-08-11 |
Family
ID=77516900
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW108145567A TWI736060B (en) | 2019-12-12 | 2019-12-12 | High-resolution video image processing method, device, and electronic device |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWI736060B (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105761267A (en) * | 2016-03-08 | 2016-07-13 | 重庆邮电大学 | Image processing method and device |
CN108139799A (en) * | 2016-04-22 | 2018-06-08 | 深圳市大疆创新科技有限公司 | The system and method for region of interest (ROI) processing image data based on user |
-
2019
- 2019-12-12 TW TW108145567A patent/TWI736060B/en active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105761267A (en) * | 2016-03-08 | 2016-07-13 | 重庆邮电大学 | Image processing method and device |
CN108139799A (en) * | 2016-04-22 | 2018-06-08 | 深圳市大疆创新科技有限公司 | The system and method for region of interest (ROI) processing image data based on user |
Also Published As
Publication number | Publication date |
---|---|
TW202123703A (en) | 2021-06-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2018028583A1 (en) | Subtitle extraction method and device, and storage medium | |
CN112954450B (en) | Video processing method and device, electronic equipment and storage medium | |
KR102087882B1 (en) | Device and method for media stream recognition based on visual image matching | |
CN109951635B (en) | Photographing processing method and device, mobile terminal and storage medium | |
CN106060470B (en) | Video monitoring method and system | |
US20220084304A1 (en) | Method and electronic device for image processing | |
CN110335216B (en) | Image processing method, image processing apparatus, terminal device, and readable storage medium | |
CN110349161B (en) | Image segmentation method, image segmentation device, electronic equipment and storage medium | |
CN103198311A (en) | Method and apparatus for recognizing a character based on a photographed image | |
CN113496208B (en) | Video scene classification method and device, storage medium and terminal | |
CN113255685B (en) | Image processing method and device, computer equipment and storage medium | |
CN112149583A (en) | Smoke detection method, terminal device and storage medium | |
CN110717452B (en) | Image recognition method, device, terminal and computer readable storage medium | |
US11348254B2 (en) | Visual search method, computer device, and storage medium | |
WO2022170982A1 (en) | Image processing method and apparatus, image generation method and apparatus, device, and medium | |
CN111401206A (en) | Panorama sharing method, system, device and medium | |
CN108769521B (en) | Photographing method, mobile terminal and computer readable storage medium | |
CN111507142A (en) | Facial expression image processing method and device and electronic equipment | |
CN110222576B (en) | Boxing action recognition method and device and electronic equipment | |
CN110349108B (en) | Method, apparatus, electronic device, and storage medium for processing image | |
TWI736060B (en) | High-resolution video image processing method, device, and electronic device | |
CN112417209A (en) | Real-time video annotation method, system, terminal and medium based on browser | |
CN110610178A (en) | Image recognition method, device, terminal and computer readable storage medium | |
US10282633B2 (en) | Cross-asset media analysis and processing | |
CN112995488B (en) | High-resolution video image processing method and device and electronic equipment |