DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1-3, 6-13 is/are rejected under 35 U.S.C. 103 as being unpatentable over Lim (PGPUB: 20180102055 A1) in view of Zhang (WO 2020228353 A1), and further in view of HUANG (CN 101673403 A).
Regarding claims 1, 10, and 11. Lim teaches an object tracking device comprising:
a memory storing instructions; and one or more processors (see Fig. 4) configured to execute the instructions to:
extract target candidates from time series images (see Fig.5, paragraph 75, the object extraction module 522 extracts objects from frames of the received image);
update a movement pattern of the target (see paragraph 67, the moving object candidate detection module 524 may store the shape of bicycle wheels and a bicycle rider on the bicycle as a pattern, determine a similarity between the object extracted by the object extraction module 522 and the pattern of the bicycle, and detect a moving object candidate).
search for and track the target using a confidence level indicating similarity with a target model among the target candidates extracted in the search range (see Fig. 5, paragraph 75 and 76, the moving object candidate detection module 524 detects a moving object candidate from the extracted objects, at ST620. When the moving object candidate is detected, the mobility determination module 526 determines similarities between the moving object candidate and the predefined patterns of a bicycle, two-wheeled vehicle and pedestrian, and determines whether the moving object candidate is moving, at step ST630. When it is determined that the moving object candidate is moving, the moving object possibility determination module 528 performs a HOG operation and SVM weight operation on the moving object candidate, and finally sets the moving object candidate to a moving object, at step ST640); and
update the target model using the target candidates extracted in the search range (see Fig. 5, paragraph 65, During the object extraction process, the object extraction module 522 may use the Canny edge detection algorithm, the line edge detection algorithm, the Laplacian edge detection algorithm or the like. The object extraction module 522 may detect a boundary line using such an algorithm, and extract an object by grouping areas distinguished from the background based on the detected boundary line).
However, Lim does not expressly teach to update a search range based on frame information of a target in a previous image in a time series.
Zhang teaches that determine the search range rectangle of the target to be tracked in the current frame of image according to the acceleration and direction calculation results; in step 300, the calculation method of the search range rectangle is specifically: taking the center position of the target to be tracked in the i+1th frame as the intersection of the diagonals of the search range rectangle (see Fig. 2, page 8, lines 14-18); Step a: Calculate the acceleration of the target to be tracked in the current frame of image according to the displacement of the previous two frames of image; Step b: Determine the search range rectangle of the target to be tracked in the current frame of image according to the acceleration calculation result; Step c: Extract the candidate frame of the target to be tracked in the current frame image along the diagonal of the rectangular frame of the search range through the RPN network, and perform feature analysis on the candidate frame to obtain the target to be tracked in the current frame image In the location (see page 12, line 4-11).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Lim by Zhang for providing determine the search range rectangle of the target to be tracked in the current frame of image according to the acceleration and direction calculation results, calculate the acceleration of the target to be tracked in the current frame of image according to the displacement of the previous two frames of image, as teach to update a search range based on frame information of a target in a previous image in a time series. Therefore, the combination of the teaching, suggestion, or motivation in the prior art would have led one of ordinary skill to modify the prior art reference or to combine prior art reference teachings to arrive at the claimed invention.
However, the combination does not expressly teach wherein the template is formed by a distribution of weights according to positions in the template and each weight corresponds to a probability of a target presence.
HUANG teaches that a target tracking method under a complex interference scene, comprising the following steps: (1) moving target detection of moving target in each entering video, determining initial position whose occurrence, using the self-adaptive gauss background model detection, and through method of morphological processing independent of the moving target area, (2) target characteristic modelling: each one motion target area detected, extracts a plurality of feature information for matching. and performing multiple feature information fusion, establishing target characteristic template, directly affect the accuracy of target tracking and robust multi-target modelling method -electrical, this invention through the multiple information of target color, gradient, texture and so on, by the nonnegative matrix decomposition method to extract and the consistency constraint, the each characteristic vector anelastic the probability density distribution function of video image, information entropy of each characteristic as an information quantity size of the metric, based on the decided feature fusion weight distribution. for establishing target characteristic template, which greatly improves the resolution of the target under complex scene, (3) matching area search in the current frame video, searching related best area with the target characteristic template in the specific region. as the current frame target tracking and positioning, the invention through multiple characteristic convergence Mean-Shift search method, using iteration method to reduce the searching optimal matching of time, at the same time, combined with the idea of simulated annealing to avoid iterative search process local optimal solution; it greatly improves the accuracy and efficiency of the method, (4) updating the target characteristic template: characteristics of the target tracking locating target characteristic template according to the characteristic information of the current frame target area and front three frames of the target characteristic template, updating the current frame target characteristic template of weights, the self-adaptive update (see page 3 and 4, lines 23-42 and 1-8).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination by HUANG for providing to extract and the consistency constraint, the each characteristic vector anelastic the probability density distribution function of video image, information entropy of each characteristic as an information quantity size of the metric, based on the decided feature fusion weight distribution. For establishing target characteristic template, which greatly improves the resolution of the target under complex scene, (3) matching area search in the current frame video, searching related best area with the target characteristic template in the specific region; and updating the target characteristic template: characteristics of the target tracking locating target characteristic template according to the characteristic information of the current frame target area and front three frames of the target characteristic template, updating the current frame target characteristic template of weights, as wherein the template is formed by a distribution of weights according to positions in the template and each weight corresponds to a probability of a target presence. Therefore, combining the elements from prior arts according to known methods and technique, such as the each characteristic vector anelastic the probability density distribution function of video image, information entropy of each characteristic as an information quantity size of the metric, based on the decided feature fusion weight distribution and updating the target characteristic template: characteristics of the target tracking locating target characteristic template according to the characteristic information of the current frame target area and front three frames of the target characteristic template, updating the current frame target characteristic template of weights, would yield predictable results.
Regarding claim 2. The combination teaches the object tracking device according to claim 1,
wherein the processor is further configured to
discriminate a category of the target in the time series images (see Lim, Fig. 5, paragraph 67, the moving object candidate detection module 524 may detect the extracted object as a moving object candidate, when the extracted object is similar to a previously stored pedestrian pattern); and
acquire a movement pattern corresponding to the category by using correspondence information of categories and movement patterns (see Lim, Fig. 5, paragraph 67, the moving object candidate detection module 524 may store the shape of bicycle wheels and a bicycle rider on the bicycle as a pattern, determine a similarity between the object extracted by the object extraction module 522 and the pattern of the bicycle, and detect a moving object candidate), and
set the acquired movement pattern as a movement pattern of the target (see Lim, paragraph 28, the step (b-2) may include detecting a vertical component of the edge of the object, and determining a similarity between the vertical component and a predefined pattern of the moving object, in order to detect the moving object candidate).
Regarding claim 3. The combination teaches the object tracking device according to claim 1,
wherein the processor is further configured to
determine the movement pattern of the target based on the time series images (see Lim, paragraph 13, detect the object as a moving object candidate when the object has an area equal to or more than a predetermined size distinguished by the edge; a mobility determination module configured to detect a motion of the moving object candidate by comparing the current frame of the image taken by the rear camera module to a previous frame before the current frame).
Regarding claim 6. The combination teaches the object tracking device according to claim 1,
wherein the processor extends the search range in a movement direction of the target (see Zhang, Fig. 4).
Regarding claim 7. The combination teaches the object tracking device according to 6,
wherein the processor contracts the search range in a direction orthogonal to the movement direction of the target (see Zhang, page 14, line 7-11, The center position of the target to be tracked in the image of the i+1th frame is taken as the intersection of the diagonals of the search range rectangle, which defines Respectively represent the horizontal and vertical coordinates of the center position of the target to be tracked, and define the starting search origin of the next frame, i+2 frame).
Regarding claim 8. The combination teaches the object tracking device according to claim 4,
wherein the template includes weights of respective positions in an area of the template (see Lim, paragraph 76, when, the moving object candidate is detected, the mobility determination module 526 determines similarities between the moving object candidate and the predefined patterns of a bicycle, two-wheeled vehicle and pedestrian, and determines whether the moving object candidate is moving, at step ST630. When it is determined that the moving object candidate is moving, the moving object possibility determination module 528 performs a HOG operation and SVM weight operation on the moving object candidate, and finally sets the moving object candidate to a moving object), and
the processor moves a center of the weights in the search range based on a movement amount of the target (see page 15, line 8-11, determining the search range rectangle of the target to be tracked in the current frame image according to the acceleration calculation result specifically includes: The center position of the tracking target is taken as the intersection of the diagonals of the search range rectangle, defining The horizontal and vertical coordinates respectively represent the center position of the target to be tracked, and define the starting search origin of the next frame, i+2 frame).
Regarding claim 9. The combination teaches the object tracking device according to 8,
wherein the processor calculates the confidence level between image features of the candidate target multiplied by the weights in the search range and the target model (see Lim, paragraph 76, when, the moving object candidate is detected, the mobility determination module 526 determines similarities between the moving object candidate and the predefined patterns of a bicycle, two-wheeled vehicle and pedestrian, and determines whether the moving object candidate is moving, at step ST630. When it is determined that the moving object candidate is moving, the moving object possibility determination module 528 performs a HOG operation and SVM weight operation on the moving object candidate, and finally sets the moving object candidate to a moving object).
Regarding claim 12. The combination teaches the object tracking device according to claim 1, wherein the movement pattern is specified depending on a direction of a movement of the target and a probability of the movement in the direction (see Lim, Fig. 5, paragraph 74, The HOG operation indicating the direction of an edge as histograms may be used when the shape of an object is not significantly changed and has a simple internal pattern and an object can be identified through the contour line of the object. For example, the moving object possibility determination module 528 calculates the direction of the edge by performing a HOG operation on a basis of a cell having a unit size of 8 pixels 8 pixels. The direction of the edge in the cell is standardized into eight angles, and expressed as histograms for the respective angles).
Regarding claim 13. The combination teaches the object tracking device according to claim 1,
wherein the movement pattern indicates one of an omni-directional type in which the target moves in any direction from a current position with almost a same probability (see Lim, paragraph 14, the moving object candidate detection module detects a vertical component of the edge of the object, and determines a similarity between the vertical component and a predefined pattern of the moving object, in order to detect the moving object candidate), a forward type in which the target moves only forward from the current position, and a forward oriented type in which the target moves forward with high probability from the current position but may also move backward (see Lim, paragraph 19, and 35, the moving object collision possibility determination unit recognizes a contour point of the large vehicle and the contour point bottom of the moving object, calculates a horizontal distance between the large vehicle and the moving object on the road based on Equation 4 below, and outputs the collision risk signal when the horizontal distance is less than a predefined horizontal danger distance; calculating a horizontal distance between the large vehicle and the moving object on the road based on Equation 4 below, and outputting the collision risk signal when the vertical distance ranges from the minimum danger distance and the maximum danger distance which are predefined in the vertical direction and the horizontal distance is less than a predefined horizontal danger distance; may include recognizing a contour point of the large vehicle and the contour point bottom of the moving object, calculating a vertical distance between the rear camera module and the moving object on the road based on Equation 2 below, calculating a horizontal distance between the large vehicle and the moving object on the road based on Equation 4 below, and outputting the collision risk signal when the vertical distance ranges from the minimum danger distance and the maximum danger distance which are predefined in the vertical direction and the horizontal distance is less than a predefined horizontal danger distance).
Claim(s) 5 is/are rejected under 35 U.S.C. 103 as being unpatentable over Lim (PGPUB: 20180102055 A1) in view of Zhang (WO 2020228353 A1), in view of HUANG (CN 101673403 A), and further in view of Liu (CN 110119730 A).
Regarding claim 5. The combination teaches the object tracking device according to claim 1,
wherein the processor determines the search range so as to correspond to a movement direction of the target (see Zhang, page 14, line 3-4, Determine the search range rectangle of the target to be tracked in the current frame of image according to the acceleration calculation result).
However, the combination does not expressly teach rotates the search range.
Liu teaches that setting the time interval taken video image frame, in the present embodiment, is set to 5s, that is, according to the time sequence of the video from each video intercepting image frame every 5s. the adjacent image frames by using the adjacent frame difference method for feature point matching, judging whether the adjacent two frame images is consistent, when present in the video of the moving object, there will be a difference in gray between the adjacent frames, calculating the two frame image gradation difference absolute value, a static object on the difference image displayed is full 0, and mobile object especially the mobile profile of the object due to the presence of the grey scale change is non-0. the characteristic image is judged, two adjacent frame image if so, abandoning the recording time of the previous image frame, if not consistent, then using the SIFT algorithm identifying extraction is not the same. scale invariant feature conversion (Scale-invariant feature transform SIFT) is a computer vision algorithm for detecting with the locality feature description in the image; It is in the space scale finding the extreme point, and extracts the position, scale and rotation invariant, its application range including object identification, robot map sensing and navigation, image stitching, 3D model building, gesture recognition, image tracking and motion comparison (see page 9, line 9-25).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination by Liu for providing that it is in the space scale finding the extreme point, and extracts the position, scale and rotation invariant, its application range including object identification, robot map sensing and navigation, as to rotate the search range. Therefore, combining the elements from prior arts according to known methods and technique, such as in the space scale finding the extreme point, and extracts the position, scale and rotation invariant, would yield predictable results.
Response to Arguments
Applicant’s arguments with respect to claim(s) 1, 10 and 11 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to XIN JIA whose telephone number is (571)270-5536. The examiner can normally be reached 9:00 am-7:30pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Gregory Morse can be reached at (571)272-3838. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/XIN JIA/Primary Examiner, Art Unit 2663
Prosecution Insights