# US 2023/0410262 A1 · Noise reduction and feature enhancement for a thermal imaging camera > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2023/0410262 A1 - Document ID: US20230410262A1 - Status: Published application - Kind: A1 - Application number: 18/337,219 - Filed: 2023-06-19 - Published: 2023-12-21 - Earliest claimed: 2022-06-20 - Assignees: Seek Thermal, Inc. - Applicants: Seek Thermal, Inc. - Inventors: Daniel Bode; Brian Nehring; Sofia Hamrin; William J. Parrish - Web page: https://patentgrep.com/patent/US20230410262A1 ## CPC - G06T 5/70: Denoising; Smoothing - H04N 23/23: from thermal infrared radiation - G01J 5/48: Thermography; Techniques using wholly visual means - G06T 5/20: using local operators - G06T 5/40: using histogram techniques - G06T 7/11: Region-based segmentation - G06T 7/13: Edge detection - G06V 10/56: relating to colour - H04N 23/81: for suppressing or minimising disturbance in the image signal generation - H04N 25/21: for transforming thermal infrared radiation into image signals - G06T 2207/10024: Color image - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION - G01J: MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY (light sources F21, H01J, H01K, H05B; investigating properties of materials by optical means G01N) - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract Thermal cameras, particularly less elaborate inexpensive designs, may deliver high temperature resolution at the expense of less clear displayed images. For many low visibility uses, such as firefighting, it may be advantageous to sacrifice some temperature detail for clarity and object recognition. Without sacrificing the actual detailed temperature information available in a captured scene, it may be beneficial to apply filters in series to raw image data that reduce visible noise and build back in feature contrast for the purposes of providing a user a displayed image that permits ease of object recognition, while still providing the raw data to a thermography module to maintain high resolution of temperature information if desired. In addition, more efficient and useful edge highlighting techniques are disclosed. For very low contrast scenes, such as smoky rooms and the like, other specific filters and equalization methods may be applied. ## Claims ### Claim 1 A method for reducing noise in a thermal image in a thermal camera including a thermal imaging sensor, and image processing capability for implementing a signal processing chain of processor executed functions, the method comprising: receiving partially processed image frame data at a noise reduction filter in the signal processing chain, the partially processed image frame data comprising frames of pixel values; and, for each pixel of a plurality of pixels of the frame: calculating n delta values for the pixel, each of the delta values comprising a difference between the pixel value of the pixel and the pixel value of a near or nearest neighbor pixel; deriving, based on a weight look up table (LUT), n weight values each corresponding to one of the n delta values; summing the n weight values; deriving n weighted delta values each equal to a product of the corresponding weight and delta values; summing the n weighted delta values; and passing forward to the signal processing chain a filtered pixel value equal to the received pixel value for the pixel minus a quotient of the sum of the n weighted delta values divided by the sum of the n weight values. ### Claim 2 (depends on 1) The method of claim 1, wherein the weight LUT contains discrete values of a function that yields diminished weight for increasing delta values. ### Claim 3 The method of claim of 2, wherein the function is a gaussian curve centered at a delta of zero. ### Claim 4 (depends on 3) The method of claim 3, further comprising determining a contrast of the scene based on a width of a histogram of the image intensity values of at least one of all or a subset of all pixels in an image frame of the scene; and varying a width of the gaussian curve depending on the width of the histogram. ### Claim 5 (depends on 1) The method of claim 1, further comprising, for each of the plurality of pixels: applying a low pass filter (LPF) to the pixel to generate a LPF pixel value; subtracting the LPF pixel value from the received pixel value of the pixel; applying a high pass filter (HPF) to the LPF pixel value to generate a HPF pixel value; applying a variable gain to the HPF pixel value according to a gain LUT; adding the LPF pixel value to the pixel value derived from the gain LUT to derive a feature enhanced pixel value; and passing forward the feature enhanced pixel value to the signal processing chain. ### Claim 6 The method of claim of 5, wherein the gain LUT boosts pixel values with values a predetermined threshold above or below zero and leaves pixel values unchanged if the pixels are within the predetermined threshold of zero. ### Claim 7 (depends on 5) The method of claim 5, wherein the feature enhancement filter follows the noise reduction filter in the signal processing chain. ### Claim 8 A method for enhancing features in a thermal image in a thermal camera including a thermal imaging sensor, and image processing capability for implementing a signal processing chain of processor executed functions, the method comprising: receiving a frame of partially processed image frame data comprising pixel values at a feature enhancement filter in the signal processing chain; and, for each pixel of a plurality of pixels of the frame: applying a low pass filter (LPF) to the pixel to generate a LPF pixel value; subtracting the LPF pixel value from the received pixel value of the pixel; applying a high pass filter (HPF) to the LPF pixel value to generate a HPF pixel value; applying a variable gain to the HPF pixel value according to a gain LUT; adding the LPF pixel value to the pixel value derived from the gain LUT to derive a feature enhanced pixel value; and passing forward the feature enhanced pixel value to the signal processing chain. ### Claim 9 The method of claim of 8, wherein the gain LUT boosts pixel values with values a predetermined threshold above or below zero and leaves pixel values unchanged if the pixels are within the predetermined threshold of zero. ### Claim 10 (depends on 8) The method of claim 8, further comprising, for each pixel of the plurality of pixels: calculating n delta values for the pixel, each of the delta values comprising a difference between the pixel value of the pixel and the pixel value of a near or nearest neighbor pixel; deriving n weight values corresponding to each delta from a weight look up table (LUT); summing the n weight values; deriving n weighted delta values each equal to a product of the corresponding weight and delta values; summing the n weighted delta values; and passing forward to the signal processing chain a filtered pixel value equal to the received pixel value for the pixel minus a quotient of the sum of the n weighted delta values divided by the sum of the n weight values. ### Claim 11 (depends on 10) The method of claim 10, wherein the weight LUT contains discrete values of a function that yields diminished weight for increasing delta values. ### Claim 12 The method of claim of 11, wherein the function is a gaussian curve centered at a delta of zero. ### Claim 13 (depends on 12) The method of claim 12, further comprising determining a contrast of the scene based on a width of a histogram of the image intensity values at least one of all or a subset of pixels in an image frame of the scene; and varying a sigma of the gaussian depending on predetermined values of the contrast. ### Claim 14 (depends on 12) The method of claim 12, further comprising determining a contrast of the scene based on a width of a histogram of the image intensity values of at least one of all or a subset of all pixels in an image frame of the scene; and varying a width of the gain LUT depending on predetermined values of the contrast. ### Claim 15 (depends on 10) The method of claim 10 wherein the feature enhancement filter follows the noise reduction filter in the signal processing chain. ### Claim 16 (depends on 10) The method of claim 10 further comprising: receiving image frame data at a colorization element and an edge highlighting filter element in the signal processing chain, the image data comprising frames of pixel values; assigning colors to each pixel corresponding to a predetermined relationship between color and at least one of pixel intensity or scene temperature of the region of the imaged scene corresponding to a given pixel, wherein the colors are assigned from a primary color look up table (LUT); applying an edge filter to the frame pixels to generate a map of edge strength values of each pixel; converting the edge strength values to a histogram of discrete edge strength bins comprising a number of edge strength bins corresponding to pixel edge strength above a predetermined value divided into a predetermined number of bins; creating a series of color LUTs each color LUT in the series comprising a shift in brightness from the primary color LUT, wherein the number of color LUTs corresponds to the predetermined number of bins, and the shift in brightness between each color LUT corresponds to a total desired shift for maximum edge strength divided by the predetermined number of bins; and replacing the colors assigned to individual pixels from the primary color LUT that have a pixel edge strength above the predetermined value, with colors from the shifted LUT corresponding to the individual pixel's edge strength bin. ### Claim 17 A method of improving image contrast for low contrast imaged scene, comprising: determining the contrast of the scene by the width of a histogram of the image intensity values of all or a subset of all pixels in an image frame of the scene; and varying the linearity of the equalization applied to the pixel values to be more linear the narrower the width of the intensity histogram.