Abstract: Disclosures of the present invention describe a moving object detection system and method, wherein a pre-processer module, a feature extraction module, an image optical flow estimation module, a feature points grouping module, and a moving object determination module are provided in a controlling and processing module of the system by a form of library, variables, or operands. Moreover, a feature difference calculation unit, a matrix establishing unit and a corner feature point acquiring unit are provided in the feature extraction module, and that is helpful for enhancing computing speed of the controlling and processing device in verifying corner feature points from image frames. Therefore, after the corner feature points are applied with a cluster labeling process, the moving object determination module can achieve motion detection of at least one object locating in a monitoring area by determining whether corner feature point groups move or not.

Abstract: Disclosures of the present invention describe a moving object detection system and method, wherein a pre-processer module, a feature extraction module, an image optical flow estimation module, a feature points grouping module, and a moving object determination module are provided in a controlling and processing module of the system by a form of library, variables, or operands. Moreover, a feature difference calculation unit, a matrix establishing unit and a corner feature point acquiring unit are provided in the feature extraction module, and that is helpful for enhancing computing speed of the controlling and processing device in verifying corner feature points from image frames. Therefore, after the corner feature points are applied with a cluster labeling process, the moving object determination module can achieve motion detection of at least one object locating in a monitoring area by determining whether corner feature point groups move or not.

Abstract: In a multi-view video coding method using a hierarchical B-picture (HBP) prediction structure, multiple image sequences in the HBP prediction structure are grouped into main-view image sequences, and auxiliary-view image sequences, each of which is configured to be disposed between corresponding two adjacent main-view image sequences. Each picture to be encoded in each main-view image sequence serves as a first target picture, and is divided into multiple macroblocks. Video data in each macroblock of the first target picture is encoded based on a first reference macroblock unit in accordance with first decision criteria. Each picture to be encoded in each auxiliary-view image sequence serves as a second target picture, and is divided into multiple macroblocks. Video data in each macroblock of the second target picture is encoded based on a second reference macroblock unit in accordance with second decision criteria.

Abstract: An underwater robot is configured to be charged in response to a charging signal from a charging dock, and includes a main body having a waterproof chamber, and an electric power unit disposed in the waterproof chamber. The electric power unit includes a rechargeable battery set storing electric power, and a charging module coupled to the rechargeable battery set and configured to charge the rechargeable battery set in response to the charging signal from the charging dock.

Abstract: In a method of filling a hole in a virtual view image warped from a two-dimensional image, the hole is detected to determine the number of pixels in the hole and to obtain position information corresponding to the hole. For each pixel in the hole, a target image block containing at least a corresponding pixel in the hole, and search image blocks corresponding to the target image block are selected from the virtual view image. Central pixels of the search image blocks are disposed on the upside, downside, right and left of a central pixel of the target image block. Those of the search image blocks detected to be available are estimated using a predetermined estimation manner to determine one of the available search image blocks as a candidate image block. All pixels in the target image block are respectively filled with all pixel values in the candidate image block.

Abstract: An underwater robot is configured to be charged in response to a charging signal from a charging dock, and includes a main body having a waterproof chamber, and an electric power unit disposed in the waterproof chamber. The electric power unit includes a rechargeable battery set storing electric power, and a charging module coupled to the rechargeable battery set and configured to charge the rechargeable battery set in response to the charging signal from the charging dock.

Abstract: In a method of filling a hole in a virtual view image warped from a two-dimensional image, the hole is detected to determine the number of pixels in the hole and to obtain position information corresponding to the hole. For each pixel in the hole, a target image block containing at least a corresponding pixel in the hole, and search image blocks corresponding to the target image block are selected from the virtual view image. Central pixels of the search image blocks are disposed on the upside, downside, right and left of a central pixel of the target image block. Those of the search image blocks detected to be available are estimated using a predetermined estimation manner to determine one of the available search image blocks as a candidate image block. All pixels in the target image block are respectively filled with all pixel values in the candidate image block.

Abstract: In a multi-view video coding method using a hierarchical B-picture (HBP) prediction structure, multiple image sequences in the HBP prediction structure are grouped into main-view image sequences, and auxiliary-view image sequences, each of which is configured to be disposed between corresponding two adjacent main-view image sequences. Each picture to be encoded in each main-view image sequence serves as a first target picture, and is divided into multiple macroblocks. Video data in each macroblock of the first target picture is encoded based on a first reference macroblock unit in accordance with first decision criteria. Each picture to be encoded in each auxiliary-view image sequence serves as a second target picture, and is divided into multiple macroblocks. Video data in each macroblock of the second target picture is encoded based on a second reference macroblock unit in accordance with second decision criteria.