工337〇4〇 ' 九、發明說明： ^ 【發明所屬之技術領域】 v 本發明係提供一種依據使用者影像控制錄影功能之錄 影系統及其方法，尤指一種利用影像辨識技術來判斷使用 者是否進入睡眠狀態或是否離開錄影系統以控制錄影功能 之錄影系統及其方法。 • 【先前技術】 由於資訊普及，一般人幾乎天天都會使用電视來獲取 資訊，例如電視新聞、連續劇、運動節目等。不過當人們 因為某些因素而無法完整收視電視節目時，例如疲累而看 ' 著電視不知不覺地睡著，或是因故離開電視機前面，便會 錯過現正觀賞之節目而造成遺憾，比方說於球赛廣告時睡 著而錯過接下來之球赛等，故如何提供一避免使用:因故 • 錯過現正收視節目資訊之機制，便為現今錄影系統人性化 設計所需努力之課題。 【發明内容】 呈本發明係提供一種依據使用者影像控制錄影功能之錄 ** 影系統及其方法，以解決上述之問題。 本發明之申請專利範圍係揭露一種依據使用者影像控 制錄影功能之錄影系統，其包含有—錄影模組，用來錄^ 6 1337040 —多媒體資料；—記憶模組，用來儲存該多媒體資料；一 影像掏取模組，用來擷取—使用者影像；—分析模组，輕 合於該影像擷取模組，該分析模_絲分析該使用者影 像’以及-控制模組’轉合於該錄影模組以及該分析模組， 用來依據該分析分析該❹者影像之結果㈣該錄影 模組是否錄製該多媒體資料。 本發明之巾請專利範圍係揭露—種錄影方法，其包含 下列步驟：⑷触—㈣者f彡像；（b)分析較用者影像； 以及(C)依據步驟(b)之分析結果決定是否錄製一多媒體 料。 【實施方式】 請參閱第1圖，第1圖為本發明依據使用者影像控制 錄影功能之一錄影系統1〇之功能方塊示意圖，錄影系統 10係可為一家用錄影系統、一電腦系統，或為一電腦裝置 與一錄影裝置所結合而成之系統。錄影系統10包含有一錄 影模組12，用來錄製一多媒體資料，錄影模組12係可為 帶式錄影裝置(Video Cassette Recorder，VCR)、數位多 功能光碟（Digital Versatile Disc，DVD)錄影裝置、數位視吼 錄放影裝置（Digital Video Recorder, DVR)、或個人錄影裝 置（Personal Video Recorder, PVR)等；一記憶模組！4，用來 儲存該多媒體資料，意即錄影模組12可將該多媒體資料錄 7 1337040 * 製至記憶模組14中；一影俸擷取模組16，用來擷取一使 . 用者影像；一分析模組18，其係可以有線或無線之方式耦 : 合於影像擷取模組16，用來分析該使用者影像，分析模組 v 18係可以硬體、軟體，或韌體之型態呈現，例如為控制器、 數位訊號處理器，或程式碼等；以及一控制模組20，耦合 於錄影模組12以及分析模組18，用來依據分析模組18分 析該使用者影像之結果控制錄影模組12是否錄製該多媒 φ 體資料，控制模組20係可以硬體、軟體，或韌體之型態呈 現，例如為微控制器或程式碼等。錄影系統10另包含有一 殼體22，錄影模組12、記憶模組14、影像擷取模組16、 分析模組18，以及控制模組20係安裝於殼體22之内，意 即所有元件皆被整合於同一台機器當中。 請參閱第2圖，第2圖為本發明利用影像辨識技術控 制錄影模組12之錄影功能之流程圖，該方法係包含下列步 •.驟： 步驟100 : 開始。 步驟102 :錄影模組12於一預定時段内錄製一緩衝多媒 體資料至記憶模組14。 步驟104 :利用影像擷取模組16擷取一使用者之該使用 者影像。 步驟106 :分析模組18分析該使用者影像於該預定時段 v 1337040 步驟108 : 内是否符合一判定規則，若是執行步驟108 ; 若否，則執行步驟114。 於記憶模組14保留該緩衝多媒體資料，且控制 模組20控制錄影模組12於該預定時段後接續 錄製該多媒體資料。 步驟110 : 利用影像擷取模組16再度擷取該使用者之該 使用者影像。 步驟112 : 分析模組18分析該使用者影像是否符合該判 定規則，若是執行步驟108 ;若否，則執行步 驟 118。 步驟114 : 控制模組20刪除記憶模組14所儲存之該緩衝 多媒體資料，且控制模組20控制錄影模組12 於下一預定時段内重新錄製下一筆緩衝多媒體 資料至記憶模組14。 步驟116 : 利用影像擷取模組16再度擷取該使用者之該 使用者影像。 步驟118 : 結束錄影模組12之錄影工作。 於此對上述步驟做一詳細之說明，首先影像擷取模組 16可被裝設於對準使用者之位置，例如對準使用者臉部之 位置，藉以擷取該使用者影像，其中可設定每間隔一段時 間便利用影像擷取模組16擷取該使用者影像。而錄影模組 12可於該預定時段内預先錄製一第一緩衝多媒體資料至記 9 1337040 憶模組14，其中該預〜 、 五分鐘等’意即錄岑：時奴可設定為任何時間長声 -緩衝多媒體資料。同： '會預先錄製五分鐘長度之= 之該使用者影像，舉例來取模組16會如使^者 用者臉部影像資料$ 影像擷取模組16可_ & 模組18便會對該使用 行之後’分析 用者影像包含該㈣者臉部f彡像^析，當該使 析該使用錢部影像資料中相對於眼㈣1模組18可分 藉以分析使用者是否處於_或睡眠讀^影像資料， 例如分析模組18可依據膚色範圍擷取出 中屬於使用者臉部影像之影像特徵資訊，舉例Μ者影像 習知之膚色理論’依據統計資料中膚色可能來說’利用 該使用者影像巾哪些像素的〇與Cb值座落^圍’判斷 Cr、Cb值範圍内’而取得臉部影像資料之影像可能之 部的大小範圍），其巾可設定該㈣影像㈣之影像337〇4〇' Nine, invention description: ^ [Technical field of invention] v The present invention provides a video recording system and method thereof for controlling video recording function according to user image, especially using image recognition technology to judge user Whether to go to sleep or leave the video system to control the video recording system and its method. • [Prior Art] Due to the popularity of information, most people use television to get information, such as TV news, serials, sports programs, etc., almost every day. However, when people are unable to fully watch TV programs due to certain factors, such as tired and watching TV, unknowingly falling asleep, or leaving the front of the TV for some reason, they will miss the program being watched and cause regret. For example, when you fall asleep during the game, you miss the next game, so how to provide a avoidance: for some reason, the mechanism of the current viewing system information is designed to meet the needs of the humanized design of today's video system. SUMMARY OF THE INVENTION The present invention provides a recording system and a method thereof for controlling a video recording function according to a user image to solve the above problems. The patent application scope of the present invention discloses a video recording system for controlling a video recording function according to a user image, which comprises a video recording module for recording 6 1337040 - multimedia data; a memory module for storing the multimedia data; An image capture module for capturing a user image; an analysis module for lightly collating with the image capture module, the analysis mode_wire analyzing the user image 'and the control module' The video module and the analysis module are configured to analyze the result of the image according to the analysis (4) whether the video module records the multimedia material. The patent application scope of the present invention discloses a video recording method comprising the following steps: (4) touching-(four) the image of the user; (b) analyzing the image of the user; and (C) determining the result of the analysis according to the step (b) Whether to record a multimedia material. [Embodiment] Please refer to FIG. 1. FIG. 1 is a schematic diagram showing the function of a video recording system according to a user's video control function. The video system 10 can be a video recording system, a computer system, or A system that combines a computer device with a video device. The video system 10 includes a video module 12 for recording a multimedia material. The video module 12 can be a Video Cassette Recorder (VCR) or a Digital Versatile Disc (DVD) video device. Digital Video Recorder (DVR), or Personal Video Recorder (PVR); a memory module! 4, for storing the multimedia data, that is, the video module 12 can make the multimedia data record 7 1337040 * into the memory module 14; a shadow capture module 16 for capturing a user. An analysis module 18 can be coupled in a wired or wireless manner: the image capture module 16 is configured to analyze the user image, and the analysis module v 18 can be hardware, software, or firmware. The display module 20 is coupled to the video module 12 and the analysis module 18 for analyzing the user according to the analysis module 18, and is configured by a controller module, a digital signal processor, or a code. The result of the image controls whether the video module 12 records the multimedia φ body data, and the control module 20 can be presented in the form of a hardware, a software, or a firmware, such as a microcontroller or a code. The video system 10 further includes a housing 22, and the video module 12, the memory module 14, the image capturing module 16, the analysis module 18, and the control module 20 are mounted in the housing 22, that is, all components. They are all integrated into the same machine. Please refer to FIG. 2, which is a flow chart of the video recording function of the video recording module 12 by using the image recognition technology in the present invention. The method includes the following steps: Step: Step 100: Start. Step 102: The video module 12 records a buffered multimedia data to the memory module 14 for a predetermined period of time. Step 104: The image capturing module 16 is used to capture the user image of the user. Step 106: The analysis module 18 analyzes the user image for the predetermined time period v 1337040. Step 108: Whether a determination rule is met, if step 108 is performed; if not, step 114 is performed. The buffered multimedia material is retained by the memory module 14, and the control module 20 controls the video module 12 to continue recording the multimedia material after the predetermined time period. Step 110: The image capturing module 16 is used to retrieve the user image of the user again. Step 112: The analysis module 18 analyzes whether the user image meets the determination rule, if step 108 is performed; if not, step 118 is performed. Step 114: The control module 20 deletes the buffered multimedia material stored in the memory module 14, and the control module 20 controls the video module 12 to re-record the next buffered multimedia data to the memory module 14 for the next predetermined time period. Step 116: The image capturing module 16 is used to retrieve the user image of the user again. Step 118: End the video recording work of the video module 12. The above steps are described in detail. First, the image capturing module 16 can be installed at a position aligned with the user, for example, at a position of the user's face, so as to capture the user image. The image capturing module 16 is conveniently used to capture the user image at intervals. The video module 12 can pre-record a first buffered multimedia material to the 9 1337040 memory module 14 during the predetermined time period, wherein the pre-~, five-minute, etc. means that the slave can be set to any length of time. Sound-buffered multimedia material. Same as: 'The user's image will be recorded in advance for five minutes. For example, the module 16 will be used to make the user's face image data. The image capture module 16 can be used. After the usage line, the analysis user image includes the (4) face image, and when the user image data is analyzed, the user can be allocated with respect to the eye (4) 1 module 18 to analyze whether the user is in the _ or The sleep reading image data, for example, the analysis module 18 can extract the image feature information belonging to the user's face image according to the skin color range, for example, the skin color theory of the user's image is known to use the use according to the skin color in the statistical data. The pixels of the image towel and the Cb value are located within the range of 'determining the range of Cr and Cb' and obtaining the size range of the image of the face image data), and the towel can set the image of the image (4)
Sil中臉部膚色部分為白色，其餘部分為黑色' i貝 分析模組18可依據所擷取出之臉部影像特徵資訊產水 睛區域影像’意即可於臉部膚色部分之白色區域中找生眼 個黑色區域之眼睛區域影像，之後再利用傳統的^】兩 處理方法對該眼睛區域影像進行影像處理， 些％像 仔出眼晴寻< 像之外部輪廓’例如可利用Sobel邊緣強化法（8处1 ’ enhancement)處理該眼睛區域影像，再利用二 edge 一值化處理法處 10 1337040 理先前所產生之影像資料，再擷取所處理之影像資料之邊 緣特徵點，最後再利用蛇模型處理法處理影像資料以產生 眼睛影像之輪廓。而由該臉部影像資料產生出該眼睛區域 影像，以及由該眼睛區域影像中擷取出眼睛影像之輪廓的 影像處理方法，可不侷限於上述之影像處理方法，於此不 再詳述。 而所得到之眼睛影像輪廓便可作為判斷使用者是否進 入瞌睡或睡眠狀態之主要依據，請參閱第3圖，第3圖本 發明所得出之眼睛影像輪廓的示意圖，由該眼睛影像輪廓 最外圍之邊緣特徵點可得出一長度L以及一寬度W，當該 眼睛影像輪廓之長度/寬度比(L/W)小於一預定值時，則代 表使用者之眼睛尚未闔上，故使用者此時並未進入瞌睡或 睡眠狀態；反之，當該眼睛影像輪廓之長度/寬度比（uw) 大於該預定值時，則代表使用者之眼睛已經闔上，故使用 者此時已進入瞌睡或睡眠狀態。而分析模組18亦可加入時 間之判斷條件，意即當判斷使用者之眼睛已經闔上超過一 預定時間，則代表使用者已進入瞌睡或睡眠狀態。 除此之外，分析模組18亦可藉由分析該使用者影像之 使用者動作影像資料來作為判斷使用者是否進入瞌睡或睡 眠狀態之依據，舉例來說可分析使用者頭部轉動之頻率或 使用者身體之傾斜角度等，藉以得知使用者是否打盹或進 11 1337040 入睡眠狀態。 當分析模組18分析使用者已進入瞌睡或睡眠狀態時會 輸出一相對應訊號至控制模組20，而控制模組20便會於 記憶模組14保留該第一缓衝多媒體資料，而不會刪除該第 一緩衝多媒體資料，且控制模組20會控制錄影模組12於 該預定時段後接續錄製該多媒體資料，以避免使用者因進 入瞌睡或睡眠狀態而錯過所欲觀賞之節目；但當分析模組 18分析使用者並未進入瞌睡或睡眠狀態時會輸出一相對應 訊號至控制模組20，此時控制模組20便會刪除記憶模組 14所儲存之該第一緩衝多媒體資料，至於刪除該第一緩衝 多媒體資料之時間點係可為該預定時段之末，且控制模組 20會控制錄影模組12於下一預定時段内重新錄製下一筆 緩衝多媒體資料（一第二緩衝多媒體資料）至記憶模組14。 也就是說，錄影模組12會於該預定時段内預先錄製緩衝多 媒體資料至記憶模組14，如預先錄製一觀看節目之第一分 鐘至第五分鐘之多媒體資料(該第一緩衝多媒體資料），當 分析模組18分析使用者已進入瞌睡或睡眠狀態時，於記憶 模組14中會繼續保留該第一緩衝多媒體資料，且控制模組 20會控制錄影模組12於第五分鐘後繼續錄製該觀看節目 之第五分鐘後之多媒體資料；但當分析模組18分析使用者 並未進入瞌睡或睡眠狀態時，控制模組20會於第五分鐘末 洗掉記憶模組14所儲存之該觀看節目之第一分鐘至第五 12 1337040 分鐘之多媒體資料（該第一緩衝多媒體資料），並於第六分 鐘至第十分鐘重新錄製該第二緩衝多媒體資料至記憶模組 14，以保持最新之緩衝多媒體資料。 之後，於錄影系統10中可繼續重複執行利用影像擷取 模組16擷取使用者之該使用者影像，以及分析模組18分 析該使用者影像於該預定時段内是否符合該判定規則之流 程，於此便不再詳述。 而當分析模組18分析使用者由瞌睡或睡眠狀態轉換為 脫離瞌睡或睡眠狀態時，意即分析模組18由分析該使用者 影像符合該判定規則轉換為分析該使用者影像不符合該判 定規則時，此時控制模組20可控制錄影模組12中止錄製 該多媒體資料之工作。之後錄影系統中可繼續重複執行 . - 步驟102後之流程，於此便不再詳述。 此外，本發明分析模組18分析該使用者影像於該預定 時段内是否符合該判定規則之準則並不侷限於判斷使用者 是否進入瞌睡或睡眠狀態，亦可為判斷使用者是否離開影 像擷取模組16之一晝面擷取範圍之狀態，當判斷使用者離 開影像擷取模組16之該晝面擷取範圍，則代表使用者無法 及時觀看節目’此時控制模組20便會於記憶模組14保留 該第一緩衝多媒體資料，且會控制錄影模組12於該預定時 13 1337040 段後接續錄製該多媒體資料，以避免使用者因離開而錯過 所欲觀賞之節目；但當判斷使用者並未離開影像擷取模組 16之該畫面擷取範圍時，控制模組20便會刪除記憶模組 14所儲存之該第一緩衝多媒體資料，且控制模組20會控 制錄影模組12於下一預定時段内重新錄製下一筆緩衝多 媒體資料至記憶模組14。再者，本發明亦可設計為判斷使 用者是否進行一特定動作來作為是否錄製該多媒體資料之 依據，例如當判斷使用者進行拍手動作時，控制模組20便 會控制錄影模組12錄製該多媒體資料，此係可提供錄影系 統10之人性化人機介面設計。至於判斷使用者離開與手部 動作等之依據，於業界已有許多不同演算法則實行之，於 此便不再詳述。 此外，為了避免分析模組18誤判使用者無法觀看錄製 節目，可另外設定若分析模組18於一倒數時間内分析該使 用者影像持續符合該判定規則時’則控制模組20控制錄影 模組12錄製該多媒體資料，其中該倒數時間係小於該預定 時段，例如可額外設定判斷使用者之眼睛是否持續闔上十 秒鐘之判定規則，以避免使用者只是瞇一下眼睛而非進入 瞌睡或睡眠狀態。另外，如有兩個以上使用者，則可設定 所有使用者皆需符合該判定規則或是大於某一特定比例之 使用者符合該判定規則時，才會啟動錄影。 14 1337040 ' 此外，請參閱第4圖，第4圖為本發明另一實施例一 . 錄影系統50之功能方塊示意圖，與前述實施例不同之處為 ' 錄影系統50包含有一第一殼體52，錄影模組12、記憶模 ' 組14，以及控制模組20係安裝於第一殼體52内；一第二 殼體54，影像擷取模組16與分析模組18係安裝於第二殼 體54内；一第一無線傳輸模組56，安裝於第一殼體52内 且耦合於控制模組20 ;以及一第二無線傳輸模組58，安裝 φ 於第二殼體54内且耦合於分析模組18，第二無線傳輸模 組58係用來以無線方式傳輸相對應分析模組18分析該使 用者影像之結果之資料至第一無線傳輸模組56，其中第二 •. 無線傳輸模組58係可運用藍芽技術(Bluetooth)之無線區域 網路協定或紅外線技術等傳輸相對應分析模組18分析該 使用者影像之結果之資料至第一無線傳輸模組56 ’以提供 控制模組20控制錄影模組12之依據。本實施例中分析模 I 組18與控制模組20之連結關係係採取無線資料傳輸之方 式，而安裝於第二殼體54内之影像擷取模組16與分析模 組18係可整合於一獨立設備之内，例如一機械人裝置，意 即影像擷取模組16與分析模組18可獨立分離於錄影模組 * 12與控制模組20之外，由於安裝於第二殼體54内之影像 . 擷取模組16與分析模組18有自由擺設的優點，故可將第 二殼體54設置於可擷取使用者影像的位置，以增加錄影系 統50影像擷取之機動性，其餘作用原理則與前述實施例相 同，於此不再詳述。 15 1337040 相較於習知之錄影裝置，本發明之錄影系統及其相關 方法係利用影像辨識技術來控制錄影裝置之錄影功能，如 此一來'便可彌補使用者因故錯過現正收視節目資訊之缺憾 (例如使用者睡著或離開），而達到人性化控制錄影裝置之 機制。 以上所述僅為本發明之較佳實施例，凡依本發明申請 專利範圍所做之均等變化與修飾，皆應屬本發明專利之涵 蓋範圍。 【圖式簡單說明】 第1圖為本發明錄影系統之功能方塊示意圖。 第2圖為本發明利用影像辨識技術控制錄影模組之錄影功 能之流程圖。 第3圖本發明所得出之眼睛影像輪廓的示意圖。 第4圖為本發明另一實施例錄影系統之功能方塊示意圖。 【主要元件符號說明】 10 錄影系統 12 錄影模組 14 記憶模組 16 影像擷取模組 18 分析模組 20 控制模組 16 1337040 22 殼體 50 錄影系統 52 第 一殼體 54 第二殼體 56 第 一無線傳輸模 組 58 第二無線傳輸模In the Sil, the skin color of the face is white, and the rest is black. The i's analysis module 18 can find the white image of the facial skin color part according to the image information of the facial image. The eye area image of the black area is imaged, and then the image processing of the image of the eye area is performed by using the conventional two methods. Some of the images are like the eye contours. For example, the outer contour of the image can be enhanced by Sobel edge. The method (8 places 1 'enhancement) processes the image of the eye region, and then uses the image processing method previously generated by the edge equalization processing method, and then extracts the edge feature points of the processed image data, and finally reuses The snake model process processes the image data to produce an outline of the eye image. The image processing method for generating the image of the eye region from the facial image data and extracting the contour of the eye image from the image of the eye region is not limited to the image processing method described above, and will not be described in detail herein. The obtained image contour of the eye can be used as the main basis for judging whether the user enters a sleepy state or a sleep state. Please refer to FIG. 3, FIG. 3, which is a schematic diagram of the outline of the eye image obtained by the present invention, from the outermost contour of the eye image contour. The edge feature point can obtain a length L and a width W. When the length/width ratio (L/W) of the eye image contour is less than a predetermined value, the user's eyes are not yet on, so the user When the length/width ratio (uw) of the contour of the eye image is greater than the predetermined value, the user's eyes are already on, so the user has entered sleep or sleep at this time. status. The analysis module 18 can also add the judgment condition of the time, that is, when it is judged that the user's eyes have been squatted for more than a predetermined time, the user has entered a doze or sleep state. In addition, the analysis module 18 can also analyze the user action image data of the user image as a basis for determining whether the user enters a sleep or sleep state, for example, the frequency of the user's head rotation can be analyzed. Or the angle of inclination of the user's body, etc., to know whether the user is snoring or entering 11 1337040 into a sleep state. When the analysis module 18 analyzes that the user has entered the sleep or sleep state, a corresponding signal is output to the control module 20, and the control module 20 retains the first buffered multimedia material in the memory module 14 instead of The first buffered multimedia material is deleted, and the control module 20 controls the video module 12 to continue recording the multimedia data after the predetermined time period to prevent the user from missing the desired program due to entering a sleepy state or a sleep state; When the analysis module 18 analyzes that the user does not enter the sleep state or sleep state, a corresponding signal is output to the control module 20, and the control module 20 deletes the first buffered multimedia data stored in the memory module 14. The time point for deleting the first buffered multimedia data may be the end of the predetermined time period, and the control module 20 controls the video module 12 to re-record the next buffered multimedia material (a second buffer) in the next predetermined time period. Multimedia data) to the memory module 14. In other words, the video module 12 pre-records the buffered multimedia data to the memory module 14 during the predetermined time period, such as pre-recording the multimedia data of the first minute to the fifth minute of the viewing program (the first buffered multimedia material). When the analysis module 18 analyzes that the user has entered a sleepy or sleep state, the first buffered multimedia data is retained in the memory module 14, and the control module 20 controls the video module 12 to continue after the fifth minute. The multimedia data after the fifth minute of the viewing of the program is recorded; but when the analysis module 18 analyzes that the user has not entered the sleep or sleep state, the control module 20 will wash the memory module 14 at the end of the fifth minute. The multimedia data (the first buffered multimedia material) of the first minute to the fifth 12 1337040 minutes of the program is watched, and the second buffered multimedia material is re-recorded to the memory module 14 from the sixth to the tenth minute to maintain The latest buffered multimedia material. Then, in the video recording system 10, the user image of the user is captured by the image capturing module 16 and the analysis module 18 analyzes whether the user image meets the determination rule within the predetermined time period. This will not be detailed here. When the analysis module 18 analyzes that the user transitions from the doze or sleep state to the doze or sleep state, the analysis module 18 converts the analyzed user image into the analysis rule to analyze that the user image does not meet the determination. In the case of the rule, the control module 20 can control the recording module 12 to stop recording the multimedia material. After that, the video system can continue to be repeated. - The process after step 102 will not be described in detail here. In addition, the analysis module 18 of the present invention analyzes whether the user image meets the criterion of the determination rule within the predetermined time period, and is not limited to determining whether the user enters a sleep or sleep state, and may also determine whether the user leaves the image. The state of one of the modules 16 is in the state of capturing the range. When it is determined that the user leaves the face capturing range of the image capturing module 16, the user cannot watch the program in time. The memory module 14 retains the first buffered multimedia material, and controls the video module 12 to continue recording the multimedia material after the 13 1337040 segment of the predetermined time to prevent the user from missing the desired program due to leaving; When the user does not leave the screen capture range of the image capture module 16, the control module 20 deletes the first buffered multimedia data stored in the memory module 14, and the control module 20 controls the video module. 12 Re-recording the next buffered multimedia material to the memory module 14 in the next predetermined time period. Furthermore, the present invention can also be designed to determine whether the user performs a specific action as a basis for recording the multimedia material. For example, when determining that the user performs a clap action, the control module 20 controls the video module 12 to record the recording. Multimedia material, which provides a humanized human interface design for the video system 10. As for the basis for judging the user's departure and hand movements, there are many different algorithms implemented in the industry, and will not be described in detail. In addition, in order to prevent the analysis module 18 from misjudged that the user cannot view the recorded program, the analysis module 18 may further set the control module 20 to control the video module if the user image continues to meet the determination rule within a countdown time. 12 recording the multimedia material, wherein the countdown time is less than the predetermined time period, for example, an additional determination rule for judging whether the user's eyes lasts for ten seconds may be set to prevent the user from just blinking his eyes instead of falling asleep or sleeping. status. In addition, if there are more than two users, the video can be started only when all users need to meet the decision rule or if the user who is greater than a certain ratio meets the decision rule. 14 1337040 ' In addition, please refer to FIG. 4 , which is a functional block diagram of the video recording system 50 according to another embodiment of the present invention. The difference from the foregoing embodiment is that the video recording system 50 includes a first housing 52 . The video module 12, the memory module 'group 14 and the control module 20 are mounted in the first housing 52; a second housing 54, the image capturing module 16 and the analysis module 18 are mounted in the second a first wireless transmission module 56 is mounted in the first housing 52 and coupled to the control module 20; and a second wireless transmission module 58 is mounted in the second housing 54 and The second wireless transmission module 58 is configured to wirelessly transmit the data of the result of the analysis of the user image by the corresponding analysis module 18 to the first wireless transmission module 56, wherein the second. The wireless transmission module 58 can transmit the data of the result of the user image to the first wireless transmission module 56 by transmitting the data of the corresponding analysis module 18 by using the wireless local area network protocol or the infrared technology of the Bluetooth technology. Providing a basis for the control module 20 to control the video module 12 . In this embodiment, the connection relationship between the analysis module group 18 and the control module 20 is wireless data transmission, and the image capturing module 16 and the analysis module 18 installed in the second housing 54 can be integrated. Within a stand-alone device, such as a robotic device, that is, the image capture module 16 and the analysis module 18 can be separated from the video module * 12 and the control module 20 independently, because it is mounted on the second housing 54. The image of the capture module 16 and the analysis module 18 have the advantage of being freely arranged, so that the second housing 54 can be placed at a position where the image of the user can be captured to increase the mobility of the image capture system 50. The rest of the operation principle is the same as the foregoing embodiment, and will not be described in detail herein. 15 1337040 Compared with the conventional video recording device, the video recording system and related method of the present invention utilize image recognition technology to control the video recording function of the video device, so that the user can be compensated for missing the current viewing program information. The shortcomings (such as the user falling asleep or leaving), and the mechanism of humanized control of the video device. The above are only the preferred embodiments of the present invention, and all changes and modifications made to the scope of the patent application of the present invention should fall within the scope of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a functional block diagram of a video recording system of the present invention. Figure 2 is a flow chart showing the use of image recognition technology to control the video recording function of the video module. Fig. 3 is a schematic view showing the contour of an eye image obtained by the present invention. 4 is a functional block diagram of a video recording system according to another embodiment of the present invention. [Main component symbol description] 10 Video system 12 Video module 14 Memory module 16 Image capture module 18 Analysis module 20 Control module 16 1337040 22 Housing 50 Video system 52 First housing 54 Second housing 56 First wireless transmission module 58 second wireless transmission mode