1231918 玖、發明說明 (發明說明應敘明：發明所屬之技術領域、先前技術、内容、實施方式及圖式簡單說明） [發明所屬雄循頁域] 本發明係關於一種分色處理再彙整結果之貨櫃號碼自動辨識方 法，特別是指一種針對貨櫃彩色影像的R(紅）、G(綠）、B(藍)三種成分個別 的灰階矩陣圖(Bit map)分別做貨櫃號碼自動辨識，再將R、G、B三種成分個別 的貨櫃號碼辨識結果予以彙整之貨櫃號碼自動辨識方法。 [先雛術] 按，現有之貨櫃號碼辯識方法多是對R、G、B三種成分個 別的灰階矩陣圖合併轉換成的整合灰階矩陣圖來做辨識。如此 一來，辨識結果可能會有誤差。 由此可見，上述習用物品仍有諸多缺失，實非一良善之設 計者，而亟待加以改良。 本案發明人鑑於上述習用方式所衍生的各項缺點，乃亟思 加以改良創新，並經多年苦心孤詣潛心研究後，終於成功研發 完成本件分色處理再彙整結果之貨櫃號碼自動辨識方法。 【發明目的】 本發明之目的即在於提供一種分色處理再彙整結果之貨櫃號碼 自動辨識方法，利用對貨櫃彩色影像的R、G、B三種成分個別的灰階矩陣圖 1231918 (Bit map)分別做貨櫃號碼自動辨識，再將R、G、b三種成分個別的貨櫃號碼辨 識結果予以彙整；一方面r、G、B三種成分個別的灰階矩陣圖中至少有一個較合 併轉換成之整合灰階矩陣圖來得適合貨櫃號碼辨識並可得到較佳的辨識結果；且 由於不同的貨櫃號碼顏色可能對應不同的貨櫃號碼編碼規則，r、G、B三種成分 個別灰階矩陣圖的貨櫃號碼辨識結果可以利用各種貨櫃號碼顏色之貨櫃號碼編 碼規則做為知識(Knowledge)來予以校正並彙整，以收到更好的貨櫃號碼自動辨 識最終結果。 【內容】 可達成上述發明目的之分色處理再彙整結果之貨櫃號碼自 動辨識方法，主要係對貨櫃彩色影像的r(紅）、G(綠）、B(藍)三種成分個別 的灰階矩陣圖(Bit map)分別做貨櫃號碼自動辨識，再將R、g、B三種成分個別 的貨櫃號碼辨識結果予以彙整。若R成分灰階矩陣圖的貨櫃號碼辨識結果為字串 Sr、G成分灰階矩陣圖的貨櫃號碼辨識結果為字串兌、B成分灰階矩陣圖的貨櫃 號碼辨識結果為字串SB，且底色或字體顏色的R成分高於一定程度以上之貨櫃 號碼的編碼規則為Cr、底色或字體顏色的G成分高於一定程度以上之貨植號碼 的編碼規則為Cq、底色或字體顏色的B成分高於一定程度以上之貨櫃號碼的編 碼規則為Cb，則可運用貨櫃號碼的編碼規則CR來驗證辨識結果Sr的正確性並在 Cr的規則範圍内校正sR使成為Sr，、運用貨櫃號碼的編碼規則& 果SG的正確性並在CG的規則範圍内校正況使成為免，、運用貨櫃號碼的編碼規 1231918 則cB來驗證辨識結果SB的正確性並在Cb _細峨正％使成騎，背 SG、SB’二者的可信錄高者即是貨櫃號碼自動辨識最終結果。 【實施方式】 清參閱圖一，本發明所提供之分色處理再囊整結果之貨植 號碼自_識方法，线 辨識模組1卜針對綠色成分灰階矩陣圖__馬自動辨職组12、針對藍色 成分灰階矩陣_貨櫃號碼自動辨識模组13、紅色成分高於一定程度以上之貨 櫃號碼的編碼規則庫2卜綠色成分高於一定程度以上之貨櫃號碼的編碼規則庫 22、藍色成分南於一定程度以上之貨櫃號碼的編碼規則庫幻、針對紅色成分灰 P雜陣圖校正_虎碼自動辨識結果並量計其可信性之模組3卜針對綠色成分 灰I1白矩陣圖枝正^櫃號碼自動辨識結果並量計其可信性之模組、針對藍色成 分灰階矩陣圖校正貨櫃號碼自動辨識結果並量計其可信性之模組33、比較器4。 、工色成刀咼於疋私度以上之貨櫃號碼的編碼規則庫21收容底色或字體顏色的 紅色成分高於-定程度以上(例如：紅色、紫色、橘色、白色)之施號碼的編碼 細、綠色成分咼於一定程度以上之貨櫃號碼的編碼規則庫22收容底色或字體 顏色的綠色成分南於一定程度以上(例如：綠色、白色)之貨櫃號碼的編碼規則、 藍色成分咼於一定程度以上之貨櫃號碼的編碼規則庫23收容底色或字體顏色的 藍色成分南於一定程度以上(例如：藍色、紫色、白色)之貨櫃號碼的編碼規則； 貨櫃號碼自動辨離組1卜貨櫃號碼自動辨識模組12與貨櫃號碼自動辨麵組 1231918 13對：k櫃彩色影像的紅色、綠色、藍色三種成分個別的灰階矩陣圖分別做貨櫃 號碼自動辨識’二者的辨識結果再由針對紅色成分灰階矩陣圖校正貨櫃號碼自動 辨識結果並量計其可信性之模組31、針對綠色成分灰階矩陣圖校正貨櫃號碼自 動辨識結果並量計其可信性之模組32、針對藍色成分灰階矩陣圖校正貨櫃號碼 自動辨識結果並量計其可信性之模組33分別根據紅色成分高於一定程度以上之 滅號碼的編碼規則庫21、綠色成分高於一定程度以上之貨櫃號碼的編碼規則 庫22、藍色成分高於一定程度以上之貨櫃號碼的編碼規則庫23來校正貨插號石馬 自動辨識的結果並量計辨識結果的可信性’三個娜號碼自動辨識校正結果的可 信性經比較器4比較後較高者即是貨櫃號碼自動辨識最終結果。 該貨櫃號碼自動辨識方法係包含下列步驟·· 步驟-：將貨櫃號碼之彩色影像處加以處理為紅 綠色或藍色之各別的灰階矩陣圖； 步驟二：將紅色、綠色或藍色之各別的灰階矩陣圖送 至針對紅色、綠色或藍色之貨榧號碼自動辨識模組進 行辨識； 步驟三：由針對紅色成份灰料_之校正辨識結果 及量計可信性模組、針對綠色成份灰階㈣圖之校正 辨識結果及量計可純m針對藍色成份灰階矩陣 圖之校正辨識結果及量訐可产u T j ^性模組，分別依據紅色 1231918 成分高於一定程度以上之編碼規則庫、綠色成分高於 一定程度以上之編碼規則庫或藍色之成分高於一定程 度以上之編碼規則庫，將分別針對紅色、綠色或藍色 之貨櫃號碼自動辨識模組所產出之辨識結果加以校 正，並量計其可信性； 步驟四：將該針對紅色成份灰階矩陣圖之校正辨識結 果及里汁可彳§性模組、針對綠色成份灰階矩陣圖之校 正辨識結果及量計可信性模組或針對藍色成份灰階矩 陣圖之校正辨識結果及量計可信性模組，所產出之可 信性的結果送至比較器； 步驟五：經由比較器比較後，其可信性較高者為最終 辨識之結果。 ，其中上述之步驟二可為：將紅色、綠色及藍色之灰階矩陣 圖送至同-個車牌自動辨識模組進行辨識，並分別產出三者之 辨識結果。 人工延炙步驟三可為 及里片1±模组，依據紅色成份高於—定程度以上之編 ^綠色成份高於—定减以上之編錢庫或藍色成份高 定程度以上之編錢庫，分別將三者之辨識結果加以校正 1231918 量計三者之可信性。 【特點及功效】 本毛明所提供之分色處理再彙整結果之貨櫃號碼自動辨識 ' /、…述弓丨δ登案及其他習用技術相互比較時，更具有下列 之優點： 本發明提供一種分色處理再彙整結果之貨櫃號碼自動辨識_ 方法，利用對貨櫃彩色影像的R、G、Β三種成分個別的灰階矩陣圖(Bit map) 分別做貨櫃號碼自動辨識，再將R、G、β三種成分個·貨櫃號碼辨識結果予以 彙整，-方面R、G、B三種成分個別的灰階矩陣圖中至少有一個較讀轉換成之 _適合貨概號碼辨識並可得到較佳的辨識結果；且由於不同的 ' 貨櫃號碼顏色可能對應不同的貨櫃號碼編碼規則，R、G、B三種成分個別灰階矩 陣圖的碰號碼辨識結果可糊用各種貨櫃號碼顏色之貨櫃號碼編碼規則做為 知識(Knowledge)來予以校正並彙整，以收到更好的貨櫃號碼自動辨識最終結果。每 上列洋細說明係針對本發明之一可行實施例之具體說明， 惟該實施例並非用以限制本發明之專利範圍，凡未脫離本發明 技藝精神所為之4效實施或變更，均應包含於本案之專利範圍 中0 綜上所述，本案不但在技術思想上確屬創新，並能較習用 物品增進上述多項功效，應已充分符合新穎性及進步性之法定 11 !231918 發明專利要件，絲法提丨巾請，_ t局核准本件發明專 利申請案，以勵發明，至感德便。 【圖式簡單說明】 請參閱以下有關本發明一較佳實施例之詳細說明及其附 圖’將可進-步瞭解本發明之技術内容及其目的功效；有關該 實施例之附圖為： 圖一為本發明分色處理再彙整結果之貨櫃號碼自動辨識方 法之架構實施例圖。 【主要部分代表符號】 11針對紅色成分灰階矩陣圖的貨櫃號碼自動辨識模組 12針對綠色成分灰階矩陣圖的貨櫃號碼自動辨識模組 13針對藍色成分灰階矩陣圖的貨櫃號碼自動辨識模組 21紅色成分高於一定程度以上之貨櫃號碼的編碼規則庫 22綠色成分高於一定程度以上之貨櫃號碼的編碼規則庫 23藍色成分高於一定程度以上之貨櫃號碼的編碼規則庫 3!針對紅色成分灰階矩陣圖校正貨櫃號碼自動辨識結果 並量計其可信性之模組 32針對綠色成分灰階矩陣圖校正貨櫃號碼自動辨識結果 並量計其可信性之模組 12 1231918 33針對藍色成分灰階矩陣圖校正貨櫃號碼自動辨識結果 並量計其可信性之模組 4比較器1231918 发明 Description of the invention (The description of the invention should state: the technical field, prior art, content, implementation, and drawings of the invention are briefly explained.] [The page domain of the invention] The invention relates to a color separation process and then summarizes the results. Method for automatic identification of container numbers, in particular, a separate gray map (Bit map) of three components of R (red), G (green), and B (blue) for the color image of the container is used to automatically identify the container number, and then Automatic container number identification method that combines the identification results of individual container numbers of three components of R, G, and B. [First embryo technique] According to the existing container number identification methods, most of the three components of R, G, and B are combined and transformed into an integrated gray matrix diagram for identification. As a result, there may be errors in the recognition results. It can be seen that there are still many shortcomings in the above-mentioned conventional articles, and they are not a good designer. They need to be improved. In view of the various shortcomings derived from the above-mentioned conventional methods, the inventor of this case has been eager to improve and innovate. After years of painstaking and meticulous research, he has finally successfully developed an automatic identification method for container numbers that has completed the color separation process and aggregated the results. [Objective of the Invention] The purpose of the present invention is to provide a method for automatic identification of container numbers of color separation processing and re-consolidation results, using individual gray scale matrix diagrams of the three components R, G, and B of the container color image 1231918 (Bit map) respectively. Automatically identify the container number, and then aggregate the individual container number identification results of the three components of R, G, and b. On the one hand, at least one of the gray scale matrix diagrams of the three components of r, G, and B is integrated and converted into an integrated gray. Can be used to identify the container number and obtain better identification results; and because different container number colors may correspond to different container number coding rules, the container number identification results of the individual gray-level matrix diagrams of the three components of r, G, and B The container number coding rules of various container number colors can be used as knowledge to be corrected and integrated to receive a better final result of automatic identification of the container number. [Content] The method for automatic identification of container numbers that can achieve the above-mentioned invention's color separation processing and re-consolidation results, mainly based on the individual gray scale matrices of r (red), G (green), and B (blue) three components of the container color image The bit map is used to automatically identify the container number, and then the individual container number identification results of the three components of R, g, and B are aggregated. If the container number recognition result of the R component grayscale matrix chart is the string Sr, the container number recognition result of the G component grayscale matrix chart is the string exchange, and the container number identification result of the B component grayscale matrix chart is the string SB, and The coding rule for a container number with an R component of a background or font color above a certain level is Cr, and the coding rule for a G number of a background or font color with a G component above a certain level is Cq, a background color, or a font color The coding rule of a container number whose B component is higher than a certain level is Cb, then the coding rule CR of the container number can be used to verify the correctness of the recognition result Sr and correct sR to become Sr within the scope of the Cr rule. The coding rules of the numbers & results in the correctness of the SG and the correction within the scope of the CG rules are avoided. The coding rules of the container number 1231918 and cB are used to verify the correctness of the identification result SB and the Cb As a result, the trusted recorder of SG and SB 'is the final result of automatic identification of the container number. [Embodiment] Please refer to FIG. 1 for the self-identification method of the product number of the color separation process and the recapture result provided by the present invention. The line identification module 1 refers to the gray-level matrix diagram of the green component. 12, against the blue component gray scale matrix _ container number automatic identification module 13, the coding rule base of container numbers with a red component above a certain level 2 and the coding rule base of container numbers with a green component above a certain level 22, A library of coding rules for container numbers with a blue component south to a certain degree or more, a correction for the red component gray P miscellaneous array correction_tiger code that automatically recognizes the results and measures its credibility. 3 For the green component gray I1 white A module that automatically identifies the result of a cabinet number and measures its credibility; a module that calibrates the result of the automatic identification of a container number and measures its credibility against the blue component grayscale matrix chart; 33, a comparator 4 . The coding rule base 21 of container numbers whose working colors are more than the degree of privateness. The red component of the background color or font color is higher than-a certain degree (for example: red, purple, orange, white). Coded rule base for container numbers with a fine code and a green component 咼 above a certain level 22 contains coding rules for container numbers whose green color of the background color or font color is greater than a certain level (eg, green, white), blue component 咼The coding rule base 23 for container numbers above a certain level contains the coding rules for the blue components of the background color or font color that are more than a certain level (for example: blue, purple, white); the container number automatic identification group 1. The container number automatic identification module 12 and the container number automatic identification group 1231918 13 pairs: the individual gray scale matrix diagrams of the red, green, and blue components of the color image of k cabinet are used for automatic identification of the container number, respectively. The results are then corrected by the module number for the red component gray scale matrix to automatically identify the results and measure its credibility. 31, for the gray component moments for the green component The module 32 that calibrates the automatic identification result of the container number and measures its credibility, and the module 33 that calibrates the automatic identification result of the container number and measures its credibility for the blue component gray scale matrix chart. Code rule base 21 for certain numbers of annihilation numbers, code rule base 22 for container numbers with green components above a certain level, code rule base 23 for container numbers with blue components above a certain level, to correct cargo insertion stones The result of the horse's automatic identification and the credibility of the meter's identification result. The credibility of the three Na number automatic identification correction results after comparison by the comparator 4 is the final result of the automatic identification of the container number. The method for automatic identification of a container number includes the following steps: Step-: Processing the color image of the container number into red, green, or blue grayscale matrix diagrams; Step 2: Red, green, or blue The respective gray scale matrix diagrams are sent to the automatic identification module for red, green or blue goods number identification; Step 3: The calibration identification result and meter credibility module for the red component gray material _ The correction identification result and quantity for the gray scale map of the green component can be pure. The correction identification result and quantity for the gray scale matrix map of the blue component can be produced. U T j ^ module can be produced according to the red 1231918 component is higher than a certain A coding rule base with a degree or more, a coding rule base with a green degree or more than a certain degree, or a coding rule base with a blue degree or more than a certain degree, will automatically identify the red, green, or blue container number respectively. The identification result of the output is corrected, and its credibility is measured; Step 4: The corrected identification result and gray scale of the red component gray scale matrix chart The correction identification result and meter credibility module for the green component gray scale matrix chart or the calibration identification result and meter credibility module for the blue component gray scale matrix chart. The result is sent to the comparator; Step 5: After comparing with the comparator, the one with higher credibility is the final identification result. Among them, the above-mentioned step two may be: sending the grayscale matrix maps of red, green, and blue to the same automatic license plate recognition module for identification, and generating the identification results of the three respectively. The third step of artificial extension can be the 1 ± module of the inner film, based on the red component higher than—a certain degree of editing ^ green component higher than—a certain amount of money or a blue component higher than a certain degree of money The library will correct the recognition results of the three, respectively, to verify the credibility of the three. [Features and effects] Automatic identification of the container number provided by the color separation processing and re-consolidation result provided by this Maoming '/, ... Shu Gong 丨 δ registration and other conventional technologies have the following advantages when compared with each other: The present invention provides a Automatic identification of container numbers for color separation processing and reassembly results_ Method, using the individual gray maps (Bit map) of the three components R, G, and B of the container color image to automatically identify the container numbers, and then R, G, and β three components · Container number identification results are aggregated,-at least one of the three gray components of the R, G, and B components is converted into at least one _ suitable for identification of the cargo number and can obtain better identification results ; And because different container number colors may correspond to different container number coding rules, the identification results of the individual gray scale matrix diagrams of the three components of R, G, and B can be filled with knowledge of the container number coding rules of various container number colors. (Knowledge) to correct and aggregate to receive better container number to automatically identify the final result. Each of the above detailed descriptions is a specific description of a feasible embodiment of the present invention, but this embodiment is not intended to limit the scope of the patent of the present invention. Any implementation or change that does not depart from the technical spirit of the present invention should be Included in the scope of patents in this case. 0 In summary, this case is not only technically innovative, but also can enhance the above-mentioned multiple effects compared with conventional items. It should have fully met the statutory requirements of novelty and advancement. , Sifa asks, please approve this invention patent application, in order to stimulate the invention, to the sense of virtue. [Brief description of the drawings] Please refer to the following detailed description of a preferred embodiment of the present invention and the accompanying drawings' to further understand the technical content of the present invention and its purpose and effectiveness; the drawings related to this embodiment are: FIG. 1 is a diagram of an embodiment of a method for automatically identifying a container number according to the color separation processing and re-assembling result of the present invention. [Representative symbols of main parts] 11 Automatic container number identification module for red component gray scale matrix chart 12 Automatic container number identification module for green component gray scale matrix chart 13 Automatic container number identification for blue component gray scale matrix chart Module 21 Coding rule base for container numbers with a red component above a certain level 22 Coding rule base for green components with a container number above a certain level 23 Coding rule library 3 for a container number with blue components above a certain level 3! Module for correcting the automatic identification of container numbers and measuring its credibility for the gray component of the red component gray scale module 32 Module for correcting the automatic identification of container numbers and measuring its credibility for the gray component of the green component 12 1231918 33 Module 4 comparator for correcting the automatic identification result of the container number and measuring its credibility against the blue component gray scale matrix diagram
1313