Abstract: A geometric approach may be used to detect objects on a road surface. A set of points within a region of interest between a first frame and a second frame are captured and tracked to determine a difference in location between the set of points in two frames. The first frame may be aligned with the second frame and the first pixel values of the first frame may be compared with the second pixel values of the second frame to generate a disparity image including third pixels. One or more subsets of the third pixels that have a value above a first threshold may be combined, and the third pixels may be scored and associated with disparity values for each pixel of the one or more subsets of the third pixels. A bounding shape may be generated based on the scoring.

Abstract: Systems and methods are disclosed that use a geometric approach to detect objects on a road surface. A set of points within a region of interest between a first frame and a second frame are captured and tracked to determine a difference in location between the set of points in two frames. The first frame may be aligned with the second frame and the first pixel values of the first frame may be compared with the second pixel values of the second frame to generate a disparity image including third pixels. One or more subsets of the third pixels that have an disparity image value about a first threshold may be combined, and the third pixels may be scored and associated with disparity values for each pixel of the one or more subsets of the third pixels. A bounding shape may be generated based on the scoring that corresponds to the object.

Abstract: A method for automatically partitioning an article into various chapters and sections is provided and applicable for a digital article. Firstly, style combinations of a plurality of paragraphs of the digital article are recognized. Then, one or more paragraph features of the paragraphs having different style combinations are calculated. The paragraph feature may be the uniform distribution of paragraphs, the font size, the average number of words, the average paragraph spacing, or the combinations thereof. Hence, in accordance with each of the paragraph features, the style combinations are ranked. Then, a weighted average value is calculated according to the ranking of each the style combinations corresponding to the corresponding paragraph feature. And, paragraphs with weighted average values ranked in the first place are selected to be a plurality of candidate partition paragraphs. Lastly, the digital article is divided into a plurality of partitions according to the candidate partition paragraphs.

Abstract: A server-based platform for text proofreading includes a network unit, a processing unit, and a storing unit. The network unit connects to a networking device to receive a document including original paragraphs with words. The processing unit connects electronically to the network unit for identifying the words from the document, converting the words into reflow paragraphs, and generating an index data based on the contrast relationship between the original paragraphs and the converted reflow paragraphs. The storing unit connects electronically to the processing unit for storing a reflow document including the reflow paragraphs. The processing unit generates a proofread webpage including a first window and a second window. When the processing unit receives a referencing command from the networking device via the network unit, the original paragraphs and the corresponding reflow paragraphs are displayed on the first window and the second window respectively according to the index data.

Abstract: A method for generating reflow-content electronic book and a website system for the same are provided. In the method, firstly, an original paragraph of a page content in a digital file is recognized. Then, an arrangement type of lines in the original paragraph is recognized, and the lines are connected to form a reflow-content paragraph based on the arrangement type, followed with calculating a recognizing confidence value corresponding to the reflow-content paragraph. Next, displaying the reflow-content paragraph in an edit interface, followed with marking the off-threshold reflow-content paragraph. Therefore, the user can check or revise the marked reflow-content paragraph in the edit interface. Last, all of the reflow-content paragraphs are saved as a reflow-content electronic book file. Accordingly, unstructured book files can be simply converted into reflow-content electronic book files, and those reflow-content paragraphs where errors might occur can be checked rapidly.