Industrial Technology Research Institute

Kual-Zheng Lee

Luo-Wei Tsai

Guo-Ruei Chen

Pang-Chan Hung

Chia-Wen Lin

Fan-Di Jou

Yung-Cheng Kao

Zhi-Hong Ling

Hui-Hsiang Kao

Hung-Ming Chen

The method for searching a video event includes: capturing a video; obtaining multiple pixels corresponding to the video; calculating a pixel difference value of each pixel by comparing frames at different timestamps in the video; dividing the video into multiple video segments, each video segment including a group of frames; determining a representative value for each pixel in each video segment according to the pixel difference value of each pixel; generating a metadata file which records the representative value for each pixel in each video segment; and searching an event in the video according to the metadata file.

Method, device, and non-transitory computer readable medium for searching video event

A video indexing method includes: analyzing trajectory information of a plurality of objects in video data and storing a plurality of pieces of object trajectory data obtained correspondingly in a storage device; determining whether the storage device satisfies a data removal condition; when the data removal condition is satisfied, performing an object trajectory analysis on the plurality of pieces of object trajectory data to select at least one target object from the objects; and extracting the object trajectory data of the at least one target object from the storage device and generating at least one video indexing image accordingly.

Video indexing method and device using the same

According to one exemplary embodiment, a method for image processing selects at least one image from a plurality of images, and stores the plurality of images into a buffer, until a storage space of the buffer reaches a criterion. When the storage space of the buffer reaches the criterion, a procedure of image processing is performed for a plurality of temporary images stored in the buffer, which includes constructing an image dictionary according to the at least an image selected, and performing an image recovery and reconstruction procedure for the plurality of temporary images in the buffer according to the image dictionary, thereby obtaining a plurality of reconstructed images.

Method and apparatus for image processing and computer readable medium

An image processing device for reducing image noise and the method thereof are provided. The method for reducing image noise includes the following steps. A filtering operation is performed on a plurality of pixel values of a first image according to a high-pass (HP) filtering mask, a first default filtering mask and a second default filtering mask to obtain a high-pass filtering image, a first low-pass (LP) image and a second LP image, respectively. A difference between the first and the second LP images is calculated to obtain a LP difference image. A pixel weighting calculation is performed according to the HP image and the LP difference image to obtain a pixel weighting mask corresponding to the first image. The first image is calculated according to the pixel weighting mask and one of the first and the second LP images to obtain a noise reducing image.

Image processing device for reducing image noise and the method thereof

A video analytics based object counting method which can obtain and process video frames from one or more video sources is proposed. By setting a variety of parameters, calculating a reference point, and a mapping table, a sampled referenced image can be constructed to obtain image pixels variation information according to these parameters. With the changed value of multiple sampling line segments and the pre-defined reference object, total object counts can be estimated by analyzing the whole number of the triggered sampling line segments and their directional states.

Method and apparatus for video analytics based object counting

A video enhancement method is disclosed. An input image is retrieved and it is determined whether an enhancement mode is implemented to the input image. If so, motion detection is implemented to the input image that only enhancement coefficients of varied pixels of the input image are recalculated and the enhancement coefficients are updated to an enhancement coefficient table. Based on the enhancement coefficient table, pixel colors of the input image are adjusted and the adjusted image is output.

Method for video enhancement and computer device using the method

A system for image processing is provided. The system includes a region of interest (ROI) module receiving video from a camera and detects a ROI(s) in a first image. A lookup table generates a value responsive to block type for a first vanishing point (VP). A labeling module identifies a point 'p' most close to the first VP, a point 'q' most remote to the first VP and a length &#x201c;h&#x201d; between &#x201c;p&#x201d; and &#x201c;q&#x201d; in each ROI(s), and generates information on p, q and h. Another lookup table generates information on p', q&#x2032; and h&#x2032;, wherein p&#x2032; is a point most close to a second VP, q&#x2032; is a point most remote to the second VP and h&#x2032; is a length between p&#x2032; and q&#x2032; in ROI(s) in the second image. A transforming module transforms ROI(s) in the first image into an ROI in the second image.

Methods and systems for image processing in a multiview video system

A system and a method of obtaining a dimension of a target object in an image comprises receiving coordinates of a number of feature points in the image, receiving coordinates of at least one reference object in the image with a known dimension, performing a calibration to adjust the coordinates of at least one of the feature points, and receiving coordinates of the target object in the image and determining the dimension of the target object based on the coordinates of the feature points. The coordinates of at least one of the feature points are adjusted to increase an accuracy in determining the dimension of the reference object.

Systems and methods for object dimension estimation

An object tracking method may include: receiving frames of data containing image information of an object; performing an object segmentation to obtain an object motion result; and using the object motion result to conduct an object tracking. In particular, the object segmentation may include: extracting motion vectors from the frames of data; estimating a global motion using the motion vectors; and subtracting the global motion from the motion vectors to generate an object motion result.

Growing community of inventors

Chiayi County, Taiwan

Kual-Zheng Lee