National Applied Research Laboratories

National Center for High-performance Computing

Fang-Pang Lin

Chien-Hao Tseng

Shi-Wei Lo

Yi-Hao Hsiao

Chung-I Huang

Yun-Te Lin

Jyh-Hong Wu

Yu-Chung Chen

Jyh-Horng Wu

Tai-Hung Chen

Hsin-Hung Lin

Whey-Fone Tsai

Jih-Sheng Chang

Ming-Chang Shieh

Chun-Ming Su

Hui-Lin Chen

Te-Lin Chung

Yung-Ching Mai

Meng-Wei Lin

Ting-Shuan Yeh

Jo-Yu Chang

Yung-Hsiang Huang

Lun-Chi Chen

Sheng-Wen Wang

Wei-Fuu Yang

The present invention relates to a method for size estimation by image recognition of a specific target using a given scale. First, a reference objected is recognized in an image and the corresponding scale is established. Then the specific target is searched and the size of the specific target is estimated according to the acquired scale.

Method for size estimation by image recognition of specific target using given scale

An image-based vehicle classification system includes a camera and an image server connected to the camera. The camera captures images of a road to result in an image stream. The image server includes a processor for receiving the image stream from the camera. For each of the images of the image stream, the processor performs image segmentation, a thinning process, an erosion process and a dilation process, and classifies, by a neural network classifier, a vehicle image part contained in the image into one of a large-size car class, a passenger car class and a motorcycle class when it is determined that the vehicle image is crossing an imaginary line set in advance in the image for counting vehicles.

Image-based vehicle classification system

The present invention discloses a distributed automatic notification method for abnormality in remote massive monitors. If a data stream is not available, a disconnect prompt stream is generated. The data stream is converted to a continuous video. Then, the continuous video is divided into a plurality of single-frame images along the time axis, and each single-frame image is divided into a plurality of image grids. A plurality of pixels are next selected from each image grid. According to a plurality of image parameter values corresponding to the plurality of pixels, a statistical value of image grid parameter is calculated. Then whether an image grid is monochromatic can be judged according to the statistical value of image grid parameter. When a plurality of continuous single-frame images are judged to be monochromatic single-frame images, the color of the monochromatic single-frame images can be further judged.

Distributed automatic notification method for abnormality in remote massive monitors

A method for estimating a number of objects is to be implemented by a processing module, and is utilized to estimate a number of a plurality of objects having elongated shapes based on an image of the objects that contains substantially-circular end faces of the objects. The method includes steps of generating a characteristic image based on the image of the objects, the characteristic image containing a plurality of characteristic patterns which are associated with the end faces of the objects, calculating a number of the characteristic patterns based on the characteristic image, and obtaining an estimated number of the objects based on at least the number of the characteristic patterns thus calculated.

Method for estimating a number of objects

In a method for monitoring water level of a water body, a monitoring system is configured to: capture a current image that has a portion of the water body, and a remaining portion aside from the portion of the water body; process the current image into a processed image that includes a water body region corresponding to the portion of the water body, and a background region corresponding to the remaining portion of the current image; mark, on the processed image, a plurality of virtual alert points according to a predetermined water level of the water body; determine whether at least one of the virtual alert points is located within the water body region of the processed image; and generate a monitoring result according to the determination thus made.

Method for monitoring water level of a water body and system for implementing the method

The present invention provide an automatic, frame-by-frame, rapid, and accurate search method for video clip. It comprises the following steps. First, acquire and store a plurality of videos to a storage unit. Each video comprises a plurality of frames, respectively. Then, a processing unit extracts the characteristics of the plurality of frames and gives a plurality of characteristic values. Next, the processing unit calculates a plurality of difference values between the characteristic value of each frame and that of the previous frame and between the characteristic value of the subsequent frame and that of the current frame. Store a data set, which is produced by corresponding the plurality of difference values between all frames of each video to the time axis, to the storage unit. Finally, the processing unit compares a data set of any video with a plurality of data sets of other videos and gives a search result.

Search method for video clip

A liquid level detection method includes capturing an image of a liquid surface, a structural surface, and graduation markings provided on the structural surface using an image-capturing device to thereby obtain an initial image. Subsequently, the initial image is processed so as to generate a processed image, and a level reference value of the liquid surface is obtained from the processed image. The level reference value represents a height of the liquid surface in terms of inherent characteristics of the processed image. Lastly, a liquid level of the liquid surface is calculated based on a relative proportional relation among the level reference value, an overall height of the processed image in terms of the inherent characteristics of the processed image, and dimensions of any one of the initial and processed images relative to the graduation markings.

Liquid level detection method

A remote control method of tile display is disclosed for a tile display system, which contains a host computer, several projector modules, and a screen. The images produced by the projector modules are simultaneously displayed on the screen as a single image. The remote control method includes the steps of: obtaining setting parameters of each projector using a control program on the host computer via a network; storing the setting parameters of each projector in a storage module of the host computer; simultaneously generating a projector parameter table for each projector; updating the projector parameter table when the parameters of any projector are modified; generating a control message for the corresponding projector according to the new parameter table; transmitting the control message to the corresponding projector via the network; and converting the control message into one recognizable by the projector in order to make corresponding modifications.

Remote control method of tile display

Growing community of inventors

Hsinchu, Taiwan

Fang-Pang Lin