Systems and methods for smoke-reduction in images

A method for smoke reduction in images comprising:

• accessing an RGB image of an object obscured by smoke, the RGB image including a plurality of pixels;
• determining a dark channel matrix of the RGB image, where the dark channel matrix includes, for each pixel of the plurality of pixels, a minimum color component intensity for a respective pixel area centered at the respective pixel;
• estimating an atmospheric light matrix for the RGB image based on the dark channel matrix, wherein the atmospheric light matrix includes an atmospheric light component value for each of the plurality of pixels, wherein the estimating the atmospheric light matrix includes, for each pixel x of the plurality of pixels:
• determining an atmospheric light component value for the pixel x as:
• A(x)=max(min(Ic(y)))*coef, for all γ∈Ω(x),
• wherein:
• Ω(x) is a pixel area centered at pixel x,
• y is a pixel of the pixel area Ω(x),
• Ic(y) is an intensity value of a color component c of the pixel y, and
• coef is a predetermined coefficient value;
• determining a transmission map based on the atmospheric light matrix and the dark channel matrix;
• de-hazing the RGB image based on the transmission map to reduce the smoke in the RGB image;
• and
• displaying the de-hazed RGB image on a display device.

A method for smoke reduction in images comprising:

• accessing an RGB image of an object obscured by smoke, the RGB image including a plurality of pixels;
• determining a dark channel matrix of the RGB image, where the dark channel matrix includes, for each pixel of the plurality of pixels, a minimum color component intensity for a respective pixel area centered at the respective pixel;
• estimating an atmospheric light matrix for the RGB image based on the dark channel matrix, wherein the atmospheric light matrix includes an atmospheric light component value for each of the plurality of pixels;
• determining a transmission map based on the atmospheric light matrix and the dark channel matrix, wherein the determining the transmission map includes determining, for each pixel x of the plurality of pixels, a transmission component value as:
• T
• ⁡
• (
• x
• )
• =
• 1
• -
• ω
• *
• I_DARK
• ⁢
• (
• X
• )
• A
• ⁡
• (
• X
• )
• ,
• wherein:
• ω is a predetermined constant,
• I_DARK(x) is the dark channel matrix value for the pixel x, and
• A(x) is the atmospheric light component value for the pixel x;
• de-hazing the RGB image based on the transmission map to reduce the smoke in the RGB image;
• and
• displaying the de-hazed RGB image on a display device.

A method for smoke reduction in images comprising:

• accessing an image obscured by smoke, the image including a plurality of pixels, where each pixel of the image includes a plurality of color components;
• for each pixel of the plurality of pixels:
• determining a dark matrix channel value for the respective pixel as a minimum color component intensity value for a respective pixel area centered at the respective pixel;
• and
• estimating an atmospheric light component value for the pixel based on the minimum color component intensity value for each pixel of the pixel area;
• de-hazing the image based on the atmospheric light component value for each of the plurality of pixels, wherein the de-hazing the image includes:
• determining a transmission map, for each pixel x of the plurality of pixels as:
• T
• ⁡
• (
• x
• )
• =
• 1
• -
• ω
• *
• I_DARK
• ⁢
• (
• X
• )
• A
• ⁡
• (
• X
• )
• ,
• wherein:
• ω is a predetermined constant,
• I_DARK(x) is the dark channel matrix value for the pixel x, and
• A(x) is the atmospheric light component value for the pixel x,
• converting the image to a YUV image;
• determining Y′, for each pixel x of the plurality of pixels as
• Y
• ′
• (
• x
• )
• =
• Y
• ⁡
• (
• x
• )
• -
• A
• ⁡
• (
• x
• )
• T
• ⁡
• (
• x
• )
• ;
• and
• replacing a Y channel of the YUV image with the determined Y′ to provide a Y′UV image;
• and
• displaying the de-hazed image on a display device.

A method for smoke reduction in images comprising:

• accessing an image obscured by smoke, the image including a plurality of pixels, where each pixel of the image includes a plurality of color components;
• for each pixel of the plurality of pixels:
• determining a dark matrix channel value for the respective pixel as a minimum color component intensity value for a respective pixel area centered at the respective pixel;
• and
• estimating an atmospheric light component value for the pixel based on the minimum color component intensity value for each pixel of the pixel area, wherein the estimating the atmospheric light component value includes, for each pixel x of the plurality of pixels:
• determining the atmospheric light component value for the pixel x as:
• A(x)=max(min(Ic(y)))*coef, for all y∈Ω(x),
• wherein:
• Ω(x) is a pixel area centered at pixel x,
• y is a pixel of the pixel area Ω(x),
• Ic(y) is an intensity value of a color component c of the pixel y, and
• coef is a predetermined coefficient value;
• de-hazing the image based on the atmospheric light component value for each of the plurality of pixels;
• and
• displaying the de-hazed image on a display device.

A system for smoke reduction in images comprising:

• a light source configured to provide light;
• an imaging device configured to acquire images;
• an imagining device control unit configured to control the imaging device, the control unit comprising:
• a processor;
• and
• a memory storing instructions which, when executed by the processor, cause the system to:
• capture an image of an object obscured by smoke, by the imaging device, the image including a plurality of pixels, where each pixel of the image includes a plurality of color components;
• access the image;
• for each of the pixels:
• determine a dark channel matrix value for the respective pixel as a minimum color component intensity value for a respective pixel area centered at the respective pixel;
• estimate an atmospheric light component value for each pixel based on the minimum color component intensity value for each pixel of the pixel area;
• and
• de-haze the image based on the atmospheric light component value for each of the pixels, wherein the instructions when de-hazing the image, further cause the system to:
• determine a transmission map, for each pixel x of the plurality of pixels as:
• T
• ⁡
• (
• x
• )
• =
• 1
• -
• ω
• *
• I_DARK
• ⁢
• (
• X
• )
• A
• ⁡
• (
• X
• )
• ,
• wherein:
• ω is a predetermined constant,
• I_DARK(x) is the dark channel matrix value for the pixel x, and
• A(x) is the atmospheric light component value for the pixel X,
• convert the image to a YUV image;
• determine Y′ as
• Y
• ′
• (
• x
• )
• =
• Y
• ⁡
• (
• x
• )
• -
• A
• ⁡
• (
• x
• )
• T
• ⁡
• (
• x
• )
• ;
• and
• replace a Y channel of the YUV image with the determined Y′ to provide a Y′UV image;
• and
• display the de-hazed image on a display device.

A system for smoke reduction in images comprising:

• a light source configured to provide light;
• an imaging device configured to acquire images;
• an imagining device control unit configured to control the imaging device, the control unit comprising:
• a processor;
• and
• a memory storing instructions which, when executed by the processor, cause the system to:
• capture an image of an object obscured by smoke, by the imaging device, the image including a plurality of pixels, where each pixel of the image includes a plurality of color components;
• access the image;
• for each of the pixels:
• determine a dark channel matrix value for the respective pixel as a minimum color component intensity value for a respective pixel area centered at the respective pixel;
• and
• estimate an atmospheric light component value for each pixel based on the minimum color component intensity value for each pixel of the pixel area, wherein the instructions when estimating the atmospheric light component of the image, further cause the system to:
• determine the atmospheric light component value for the pixel x as:
• A(x)=max(min(Ic(y)))*coef,
• wherein:
• y is a pixel,
• Ic is a color component of the pixel y, and
• coef is a predetermined coefficient;
• and
• de-haze the image based on the atmospheric light component value for each of the pixels;
• and
• display the de-hazed image on a display device.