公开(公告)号：MY172797A

公开(公告)日：2019-12-12

申请号：MYPI2014003462

申请日：2014-12-15

Applicant: MIMOS BERHAD

Inventor： DR HUM YAN CHAI ,  LIANG KIM MENG ,  DR HON HOCK WOON ,  HAMAM MOKAYED ,  KELVIN LO YIR SIANG

IPC: G06K9/00

Abstract: A system (100) and method for extracting and recognizing characters which are inscribed in a vehicle license plate image has been disclosed. The system (100) comprises an image analysing processor (102) for extracting the characters by binarizing the image in an iterative manner with a convergence property which prevents over enhancing of the binarized plate image. The processor (102) includes a binarization module (106) which selectively strengthens and weakens the license plate image's pixel intensity and thereby replaces them with any one of two distinct intensity values in the iterative manner till the binarized image reaches a converging state having an upper part with high intensity value, and a lower part with low intensity value and middle part with intensities that are unattainable to convert into high intensity value.

公开(公告)号：MY188016A

公开(公告)日：2021-11-09

申请号：MYPI2014003278

申请日：2014-11-21

Applicant: MIMOS BERHAD

Inventor： ZULAIKHA KADIM ,  LIANG KIM MENG ,  DR HON HOCK WOON ,  KHAIRIL HAFRIZA BIN ZAKARIA ,  HASMARINA BINTI HASAN ,  KHARUNNISA ,  YUEN SHANG LI

IPC: G06T7/00 ,  G06T7/20

Abstract: There is disclosed a system for detecting an unattended or abandoned object within an area; the system comprises: at least one video acquisition unit (101) adapted for capturing at least one image of the area; an event analysis unit (102) being operably connected to the video acquisition unit (101); and a video display unit (109) being operably connected to the even analysis unit (102) for displaying all outputs from the event analysis unit (102). In one embodiment, the event analysis unit (102) is adapted to construct a background based on the image; detect at least one motion area within the image; detect at least one static object within the image; classify the object as to whether the detected static object is an abandoned object; and update the background based on the outputs from the all modules to obtain a final output image. A method thereof is also disclosed herein.

公开(公告)号：MY181263A

公开(公告)日：2020-12-21

申请号：MYPI2015000396

申请日：2015-02-13

Applicant: MIMOS BERHAD

Inventor： ZULAIKHA BINTI KADIM ,  KHAIRUNNISA MOHAMED JOHARI ,  DR HON HOCK WOON ,  KEE HONG SHENG

Abstract: A system (100) and method for facilitating content-based video analysis and retrieval has been disclosed. The system (100) includes a video content analysis module (104) which facilitates in extracting at least two-level of video content information from image sequences of real-time and offline video feeds. The extracted video content information is annotated to corresponding frame(s) in the image sequences and stored in a video content database (110). The video content database (110) only holds the annotated image frames acquired from the videos which can be retrieved on querying the database (110). The system (100) by employing the two-level extraction technique enables users to input complex and multi event related queries for extraction of specific frames from videos that correspond to the given query, with accuracy and speed. Thereby, the system (100) eliminates the need for re-analysing the videos at a later stage for searching or detecting specific events.

公开(公告)号：MY169565A

公开(公告)日：2019-04-22

申请号：MYPI20091829

申请日：2009-05-06

Applicant: MIMOS BERHAD

Inventor： DR HON HOCK WOON

Abstract: A method to expedite image processing by consolidating complex multiple-map based nonlinear transformation into a single solution using a single map as lookup table. The complex, multiple cascading transformations (such as the typical circular-to-rectangular correction, horizontal correction and vertical correction) are hedged off to be processed offline so that these processing time-intensive steps may be performed without affecting the online process. Only the final result from these offline transformation processes tabulated into a Master Map to be used a lookup table in a single transformation in the online process. Our method may be implemented solely as a software, including as one or more applications capable of being executed in a image-processing hardware and system. Hence, our invention may be embodied as a software or firmware updated version of an existing hardware or apparatus incorporating our aforedescribed method or software.

公开(公告)号：WO2014038924A2

公开(公告)日：2014-03-13

申请号：PCT/MY2013000154

申请日：2013-09-05

Applicant: MIMOS BERHAD

Inventor： ZULAIKHA BINTI KADIM ,  NORSHUHADA BINTI SAMUDIN ,  DR HON HOCK WOON

CPC classification number: G06T7/277 ,  G06T2207/20081 ,  G06T2207/30232

Abstract: The present invention relates to a novel method for producing a background model based on the images acquired from a non-static camera. Rule-based statistical multiple image analytical method to process multiple non-overlapping background region of a scene using predetermined threshold values to identify the preferred intensity variance for modeling the background pixel. One of the advantages of the a method for modeling dynamic scene using region-based adaptive statistical learning to model dynamic background within one camera view and scene-based modeling to model multiple non-overlapping regions of the background image scene is that it provides an accurate background representation as compared to background modeling using pixel-wise scalar value. Another advantage of the method for the present invention is that a less computational cost is used as compared to background modeling using pixel-wise statistical or kernel density. Furthermore, the method for the present invention has higher sensitive foreground detection as compared to privacy mask concept (masking out dynamic region from being modeled.

Abstract translation: 本发明涉及一种基于从非静态摄像机获得的图像来制作背景模型的新颖方法。 基于规则的统计多图像分析方法，使用预定阈值来处理场景的多个非重叠背景区域，以识别用于建模背景像素的优选强度方差。 使用基于区域的自适应统计学习建模动态场景的一种方法的一个优点是在一个摄像机视图和基于场景的建模中对动态背景进行建模，以模拟背景图像场景的多个非重叠区域，其提供了准确的 与使用像素级标量值的背景建模相比，背景表示。 本发明的方法的另一个优点是与使用像素统计或核密度的背景建模相比，使用较少的计算成本。 此外，与隐私掩码概念相比，本发明的方法具有更高的灵敏度前景检测（掩蔽动态区域被建模。