DETAILED ACTION
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claims 1-15 are pending.
Claim Rejections - 35 USC § 103
The following is a quotation of pre-AIA 35 U.S.C. 103(a) which forms the basis for all obviousness rejections set forth in this Office action:
(a) A patent may not be obtained though the invention is not identically disclosed or described as set forth in section 102 of this title, if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains. Patentability shall not be negatived by the manner in which the invention was made.
Claim(s) 1-7, 9, 11-12 and 14-15 is/are rejected under 35 U.S.C. 103 as being unpatentable over Dvir et al (US10943114B2) in view of Mouret et al (Unsupervised crop anomaly detection, 2020).
(Note: prior art reference Dvir et al (US10943114B2) is from the IDS filed on 4/23/2026)
Regarding claims 1, 14 and 15, Dvir teaches a method of identifying a soil-borne pathogen on a target crop in an agricultural parcel, comprising the steps of:
(Dvir, "For agricultural purposes, for example, the current invention implements automatic systematics to identify pests, diseases, and vegetation vigor and other aspects of the crops.", [c7:5-10]; a method of identifying pests or diseases (such as a soil-borne pathogen) on a crop in an agricultural parcel)
obtaining a first digital image of the agricultural parcel in a first crop cycle, wherein in the first crop cycle the target crop is grown in the agricultural parcel;
(Dvir, "capturing visual and/or hyper-spectral or multi-spectral imagery", [c10:15-20]; Mouret, "One season of wheat crops (2016/2017) and one season of rapeseed crops (2017/2018) are considered", [Section 2.1, p5]; Dvir teaches obtaining digital images of an agricultural parcel. Mouret teaches obtaining imagery for a specific crop growing season (first crop cycle) where a specific crop is grown. Together Dvir and Mouret teach obtaining a first digital image of the agricultural parcel during a first crop cycle where a target crop is grown)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention was made to incorporate the teachings of Mouret into the system or method of Dvir in order to monitor field conditions over specific seasonal crop cycles. The combination of Dvir and Mouret also teaches other enhanced capabilities.
The combination of Dvir and Mouret further teaches:
obtaining a reference digital image of the agricultural parcel in a reference crop cycle, wherein in the reference crop cycle a reference crop is grown in the agricultural parcel and wherein the reference crop is different from the target crop;
(Dvir, " acquire the imagery from the closest possible distance", [c8:45-50]; Mouret, "historical modeling can be challenging because of crop rotation.", [Section 1, p4]; "One season of wheat crops (2016/2017) and one season of rapeseed crops (2017/2018) are considered to show that the proposed method is robust to changes in crop types.", [Section 2.1, p5]; Dvir teaches obtaining additional reference images of the agricultural parcel. Mouret teaches obtaining images of the parcel during a different growing season (reference crop cycle) where a different crop type is grown due to crop rotation. Together Dvir and Mouret teach obtaining a reference digital image of the parcel in a reference crop cycle with a reference crop different from the target crop. Incorporating Mouret into Dvir would track field health and isolate crop-specific issues across different rotational planting cycles.)
computing a first vegetation index related to a first pixel in the first digital image,
(Dvir, "The input for the calculations is a matrix of NDVI values ranging from -1 to 1, as acquired by a multi-spectral airborne camera.", [c10:60-65]; computing a vegetation index (NDVI) related to pixels in the obtained digital image)
determining a first signed distance between the first pixel and surrounding pixels thereof based on the first vegetation index, and
(Dvir, "detect regions in the digital images that differ in properties, such as brightness or color, compared to surrounding regions.", [c6:30-35]; Mouret, "Let k-dist( ) be the distance between the object and its k-th nearest neighbor.", [Section 3.4, p13]; Dvir teaches detecting regions that differ in properties compared to surrounding pixels/regions based on index values. Mouret teaches calculating the spatial or reachability distance between an object (pixel/instance) and its nearest neighbors to find anomalies. Together Dvir and Mouret teach determining a distance (which mathematically encompasses a signed difference/distance) between a first pixel and surrounding pixels based on the vegetation index. Incorporating Mouret into Dvir would quantify the exact localized numerical divergence of a pixel relative to its neighbors)
detecting a first anomaly of the first pixel if the first signed distance is below a predefined threshold;
(Dvir, "manifested in NDVI values which are on average lower by 15% to 20% than the optimal NDVI value", [c5:30-35]; "If the correlation measure exceeds a pre designed threshold value, pest detection is flagged.", [c14:35-45]; detecting an anomaly (pest/stress) when the calculated parameter difference/distance falls below or exceeds a predefined threshold level compared to the optimal or normal surroundings)
defining a reference anomaly for a reference pixel of the reference digital image; and
(Dvir, "defining the areas of interest in the images which call for more detailed examination", [c10:20-30]; Mouret, "detect anomalies in the rapeseed and wheat parcels.", [Section 3.6, p21]; applying the same anomaly definition and detection procedures to the reference digital images to define reference anomalies in the alternate crop cycle)
identifying the soil-borne pathogen of the first pixel in case the first anomaly does not match the reference anomaly.
(Dvir, "comparing them to shapes in a database in order to identify the object", [c2:35-40]; Mouret, "detect the most abnormal parcels within a season rather than detecting inter-annual abnormalities.", [Section 1, p4]; Dvir teaches identifying pathogens/pests by comparing the anomaly features. Mouret teaches distinguishing seasonal/intra-annual anomalies from inter-annual historical data (crop rotation). By combining Dvir and Mouret, one monitors the parcel across different crops; if an anomaly (disease) appears in the target crop but does not match/appear in the reference non-host crop, the system identifies a crop-specific soil-borne pathogen. Incorporating Mouret into Dvir would accurately diagnose crop-specific diseases by observing whether the anomaly persists or changes across different rotational host/non-host crops.)
Regarding claim 2, the combination of Dvir and Mouret teaches its/their respective base claim(s).
The combination further teaches the method according to claim 1, wherein defining the reference anomaly comprises the steps of:
computing a reference vegetation index relating to a reference pixel in the reference digital image, determining a reference signed distance between the reference pixel and surrounding pixels thereof, and detecting the reference anomaly of the reference pixel if the reference signed distance is below the predefined threshold.
(Dvir, "The algorithm constructs a binary map of the NDVI image", [c11:5-10]; Mouret, "compares the local density of each instance with that of its k-nearest neighbors", [Section 3.4, p13]; Dvir teaches computing a vegetation index (NDVI) for pixels in an image to construct a reference map and detect localized irregularities/anomalies. While Dvir identifies anomalous pixels using absolute index ranges, Mouret teaches improving point anomaly detection by comparing an instance to its "k-nearest neighbors" to evaluate local deviations. It would have been obvious to a person of ordinary skill in the art to enhance Dvir's pixel-level NDVI analysis with Mouret's localized neighbor-comparison technique to better detect anomalies relative to their immediate surrounding background rather than a global average. Furthermore, implementing this local comparison using a "signed distance" (i.e., a directional difference) would have been an obvious mathematical design choice to specifically isolate negative deviations (e.g., a pixel's index being lower than its neighbors, indicating localized plant stress) that fall below a predefined threshold)
Regarding claim 3, the combination of Dvir and Mouret teaches its/their respective base claim(s).
The combination further teaches the method according to claim 1, further comprising:
obtaining a second digital image of the agricultural parcel in a second crop cycle,
wherein in the second crop cycle the target crop is grown in the agricultural parcel, and
computing a second vegetation index indicative of vegetation vigour of the agricultural parcel for a second pixel in the second digital image,
determining a second signed distance between the second pixel and surrounding pixels thereof, and
detecting a second anomaly of the second pixel if the second signed distance is below the predefined threshold.
(Dvir, "assist the farmer in deciding what steps he needs to take, immediately or in the short term, or in the long term.", [c6:45-55]; Mouret, "historical modeling can be challenging because of crop rotation.", [Section 1, p4]; Dvir and Mouret combined teach long-term crop monitoring over multiple crop rotations. It is obvious to obtain a second digital image of the target crop in a subsequent target growing cycle (second crop cycle) and perform the same vegetation index and distance anomaly detection to track the long-term status of the identified pathogen)
Regarding claim 4, the combination of Dvir and Mouret teaches its/their respective base claim(s).
The combination further teaches the method according to claim 1, further comprising:
determining the first signed distance by comparing the first vegetation index with an average and standard deviation of vegetation indices of the surrounding pixels of the first pixel.
(Mouret, "defines the statistical notion of an outlier as an object deviating more than a given times the standard deviation from the mean", [Section 3.4, p14]; determining the anomaly or distance by comparing the pixel/object metric to the mean (average) and standard deviation of its local surroundings)
Regarding claim 5, the combination of Dvir and Mouret teaches its/their respective base claim(s).
The combination further teaches the method according to claim 3, further comprising:
determining the reference signed distance by comparing the reference vegetation index with an average and standard deviation of vegetation indices of the surrounding pixels of the reference pixel, and/or determining the second signed distance by comparing the second vegetation index with an average and standard deviation of vegetation indices of the surrounding pixels of the second pixel.
(Mouret, "defines the statistical notion of an outlier as an object deviating more than a given times the standard deviation from the mean", [Section 3.4, p14]; As established for Claim 4, Mouret teaches determining distance using standard deviation from the mean. Applying this identical mathematical logic to the reference image and the second image is a predictable and obvious repetitive application of the known algorithm to ensure consistent multi-cycle measurements)
Regarding claim 6, the combination of Dvir and Mouret teaches its/their respective base claim(s).
The combination further teaches the method according to claim 3, further comprising:
creating a first, a reference and a second signed distance map including signed distances of a plurality of the first, the reference and the second pixels, respectively.
(Dvir, "constructs a binary map of the NDVI image", [c11:5-10]; Mouret, "an anomaly score is given for each instance analyzed", [Section 3.4, p16]; Dvir teaches constructing metric maps based on indices, and Mouret teaches generating distance/anomaly scores for all analyzed pixels/instances. Combining Dvir and Mouret to create a distance map aggregating the signed distances for the first, reference, and second pixels is a standard, predictable mapping of the generated data array)
Regarding claim 7, the combination of Dvir and Mouret teaches its/their respective base claim(s).
The combination further teaches the method according to claim 6,
wherein the first, the reference and the second signed distance map comprise a plurality of the first, the reference and the second crop cycles, respectively, and
wherein the plurality of the first, the reference and the second crop cycles alternate with each other.
(Mouret, "historical modeling can be challenging because of crop rotation.", [Section 1, p4]; "One season of wheat crops (2016/2017) and one season of rapeseed crops (2017/2018) are considered", [Section 2.1, p5]; monitoring across multiple years involving alternating crop rotation. Structuring the distance maps such that they cover alternating target and reference crop cycles (e.g., target, then reference, then target) is an obvious reflection of standard agricultural rotation monitoring)
Regarding claim 9, the combination of Dvir and Mouret teaches its/their respective base claim(s).
The combination further teaches the method according to claim 6, further comprising:
identifying the soil-borne pathogen of the first pixel of the target crop in case the first anomaly does not exist in the reference signed distance map.
(Dvir, "comparing them to shapes in a database in order to identify the object", [c2:35-40]; Mouret, "historical modeling can be challenging because of crop rotation", [Section 1, p4]; Dvir teaches identifying targets by comparing anomaly data against references. Mouret teaches monitoring fields across different crop rotations. It is a well-known agricultural principle that many soil-borne pathogens are host-specific and are disrupted by crop rotation. Therefore, it would be obvious to a person of ordinary skill to compare the generated distance maps from the two different crop cycles. Specifically, checking that a localized anomaly exists in the target crop's map but is absent in the reference (non-host) crop's map is a predictable, standard diagnostic deduction used in agriculture to positively identify a host-specific soil-borne pathogen)
Regarding claim 10, the combination of Dvir and Mouret teaches its/their respective base claim(s).
The combination further teaches the method according to claim 9, further comprising:
identifying the first anomaly being a recurrent soil-borne pathogen in case the first anomaly exists in the second signed distance map, or identifying the first anomaly being a non-recurrent soil-borne pathogen in case the first anomaly does not exist in the second signed distance map.
(Mouret, "detecting inter-annual abnormalities.", [Section 1, p4]; Dvir teaches identifying the specific pest ([c7:5-10]). Mouret teaches analyzing data across multiple years (inter-annual). It is obvious to check if the confirmed pathogen anomaly reappears in the second target crop cycle's map to identify it as a recurrent inter-annual pathogen, yielding predictable multi-season diagnostic results)
Regarding claim 11, the combination of Dvir and Mouret teaches its/their respective base claim(s).
The combination further teaches the method according to claim 1, further comprising:
changing the predefined threshold for adjusting a level of the soil-borne pathogen to be identified.
(Dvir, " Such threshold setting can be set to match agricultural spraying policies", [c13:20-25]; Mouret, "select the percentage of anomaly to be detected by sorting the instances", [Section 3.4, p16]; changing or adjusting the predefined detection threshold to tune the sensitivity or level of the pest/pathogen identified. This is a known and obvious technique to control detection rates and match operational thresholds)
Regarding claim 12, the combination of Dvir and Mouret teaches its/their respective base claim(s).
The combination further teaches the method according to claim 1, further comprising:
isolating the first pixel of the target crop with pixels having signed distances below the predefined threshold, wherein the predefined threshold corresponds to a probability indicative of a false alarm.
(Dvir, "very few pixels cannot be ignored and even small blobs are of significance", [c7:50-55]; Mouret, "its probabilistic extension Local Outlier Probabilities (LoOP)", [Section 3.4, p13]; "provide an anomaly score belonging to the interval [0,1]", [Section 3.4, p13]; Mouret teaches using probabilistic extensions to output an anomaly score between 0 and 1, which represents the probability of an anomaly versus a normal instance (false alarm). Combining Dvir and Mouret makes isolating these pixels based on probability-driven thresholds an obvious optimization)
Regarding claim 13, the combination of Dvir and Mouret teaches its/their respective base claim(s).
The combination further teaches the method according to claim 1, wherein the soil borne pathogen is a nematode stress, wherein the target crop is soybean, and wherein the reference crop is non-host crop.
(Dvir, "a bank of potential pests/diseases, which are relevant to that specific type of crop", [c9:25-30]; Mouret, "historical modeling can be challenging because of crop rotation.", [Section 1, p4]; Dvir and Mouret combined teach a general automated system for tracking crop-specific pests across different crop rotations. Claim 13 merely restricts this generic method to a single, ordinary real-world example: a common crop (soybean), a common pest (nematode), and a standard farming practice (rotating to a non-host crop the pest dislikes). Applying a known, generic pest-detection system to a highly common, commercially standard crop and pest combination is an obvious application of a known technique. It requires no inventive step to simply point the camera at a soybean field rather than a wheat field, yielding entirely predictable diagnostic results)
Allowable Subject Matter
Claim(s) 8 is/are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening Claim(s).
The following is a statement of reasons for the indication of allowable subject matter:
Claim(s) 8 recite(s) limitation(s) related to computing crop stress states using mean operators and updates signed distance maps using minimum operators. There are no explicit teachings to the above limitation(s) found in the prior art cited in this office action and from the prior art search.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JIANXUN YANG whose telephone number is (571)272-9874. The examiner can normally be reached on MON-FRI: 8AM-5PM Pacific Time.
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, Amandeep Saini can be reached on (571)272-3382. 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.
/JIANXUN YANG/
Primary Examiner, Art Unit 2662 6/12/2026
Prosecution Insights