153 results about "Image averaging" patented technology

Determining if an input image is a foggy image includes determining an average luminance gray level of the input image, performing Sobel image processing on the input image to generate a Sobel image of the input image when the average luminance gray level of the input image is between a first image average luminance and a second image average luminance, determining a first normalization value and a second normalization value of the input image, determining a mean value and a standard deviation of the Sobel image when the first normalization value and the second normalization value are less than a first threshold value, and determining the input image as a foggy image when a sum of the mean value and the standard deviation is less than a second threshold value.

A camera automatic exposure method and a control device, the method comprising the following steps: a. obtaining the average brightness value of the current frame image; b. comparing the current frame image average brightness value with a preset average brightness value, if the current frame image If the average brightness value of the current frame image is within the allowable range of the preset average brightness value, maintain the exposure time in the current exposure setting and return to step a, otherwise continue; c. If the average brightness value of the current frame image is greater than the preset average brightness value within the allowable range, adjust the shortening of the exposure time to obtain the new exposure time in the exposure setting; if the average brightness value of the current frame image is less than the allowable range of the preset average brightness value, then adjust the lengthening of the exposure time to obtain the exposure setting a new exposure time; and e, setting the image sensor according to the new exposure time. The present invention can realize automatic exposure at low cost, simply and quickly.

The invention relates to an automatic exposure control method and device based on an FPGA and suitable for space exploration imaging. The method includes the steps that the current-frame image average brightness is acquired through the main current-frame image average brightness, time of exposure is adaptively selected according to whether a difference value between the current-frame image average brightness and the expected image average brightness is positive or negative and is large or small, continuous frames are adopted for continuous loop iteration, and therefore the current-frame image average brightness reaches the expected image average brightness. The automatic exposure control device for conducting the automatic exposure control method comprises an optical lens, an APS sensor and the exposure control FPGA. According to the method and device, the technical problems existing in excessive exposure of a target image, segmentation of the target image and balance between the automatic exposure speed and stability in the existing space exploration imaging technology are solved, the image contrast ratio is increased, faster, more stable and more accurate automatic exposure control can be conducted in various space exploration light environments, and adaptive ability in the space exploration environment imaging process is improved.

The invention provides an automatic aperture adjusting device and method based on self-adaptive adjustment. The automatic aperture adjusting device comprises an automatic aperture lens, an average brightness calculation unit, an automatic aperture control unit and an automatic aperture motor, wherein the automatic aperture lens is used for collecting images; the average brightness calculation unit is used for calculating the average brightness of an image acquired by the automatic aperture lens; the automatic aperture control unit is used for generating a control signal which controls the size of the aperture according to the difference of the average brightness calculated by the average brightness calculation unit and target brightness; and the automatic aperture motor is used for controlling the size of the aperture of the automatic aperture lens according to the control signal outputted by the automatic aperture control unit. The size of the aperture can be adaptively controlled byanalyzing the current state of the aperture, so that the device can be applied to various automatic aperture lenses, and avoids the problems such as manually adjusting parameters and the like.

The invention relates to a method for carrying out real-time identification and targeted spraying on cotton field weedsVisual system hardwares, comprising an additional light source, a CCD color video camera, an A / D converter, a DSP image processor, a D / A converter and an LCD (liquid crystal display) display; the additional light source can improve the situation in which light is not uniform or bright enough under natural conditions.The identification method comprises the following steps: a video camera installed at the front part of a weeding machine captures cotton field images in real time; after A / D conversion, the images are stored in a chip memory for being called by a DSP (Digital Signal Processor); the DSP extracts frame images from video streams and processes the frame images; green plants from soil backgrounds are segmented by the Cr characteristic method; the images are equally divided into four lines according to the arrangement of spray nozzles with each spray nozzle corresponding to a line; position features are used to recognize weeds among the lines; the weeds are recognized according to area and form features of blades; weeds position information is transmitted to a microcontroller; and the spray nozzles are opened to carry out accurate targeted spraying of weed killer to save pesticides and reduce pollution to the environment.

The invention relates to an automatic exposure control method based on adaptive expected image average brightness and suitable for space exploration imaging. The method includes the steps that firstly, a current-frame shot image is segmented through an adaptive threshold value, a main current-frame shot image is identified, and the current-frame shot image average brightness is acquired through the main current-frame shot image average brightness; secondly, histogram equalization is conducted on the identified main current-frame shot image, so that the expected image average brightness is acquired after histogram equalization; thirdly, time of exposure is selected according to whether a difference value between the current-frame shot image average brightness and the expected image average brightness is positive or negative and is large or small, continuous frames are adopted for continuous loop iteration, and therefore the current-frame shot image average brightness reaches the adaptive expected image average brightness acquired after histogram equalization. According to the method, the good exposure effect can be acquired under various conditions of a dark background and a bright target in the space exploration process, the uncertain size of the target and the like, and adaptive ability in the space exploration environment imaging process is improved.

The invention discloses a lab-space-based detection method based on image color cast. The method includes the steps of firstly, acquiring a current video frame; secondly, converting RGB (red, green, blue) into a lab space; thirdly, evaluating ratio K of image average chroma D to chroma center distance M according to a lab value; fourthly, evaluating histograms of a and b, calculating a factor according to the histograms, and multiplying the factor to the ratio K to acquire a result f; and fifthly, determining color cast occurs in an image to be detected when value of f is larger than a threshold value. Limitation of scene or prior knowledge is avoided, and compared with the prior art, the lab-space-based detection method based on image color cast is higher in detection precision and has excellent detection capacity in terms of scene color occupying large area also, universal adaptability and higher detection accuracy for color cast and reliability. In addition, conversion of color space and knowledge of the histograms are utilized flexibly, calculation is simplified, and real-time performance is high.

The invention discloses an edge extraction method based on a fusion strategy, and belongs to the field of computer application, such as image processing, pattern recognition and vision computation. The edge extraction method comprises the following steps of inputting a grayscale image; integrating extraction results of three typical edge extraction algorithms to obtain voting weight reflecting and belonging to edge possibility degree; analyzing a difference value of the maximum luminance difference and the minimum luminance difference of pixel points and a neighborhood to obtain difference weight describing luminance mutation degree; counting deleted neighborhood variance distribution, and according to the characteristic that an edge point at least has the relative larger luminance dispersity on the basis that the luminance dispersity of four neighborhoods of the whole image is average, obtaining edge distribution weight of all the pixel points; fusing three weight matrixes to carry out edge decision-making; outputting an edge image. The edge extraction method based on the fusion strategy, which is disclosed by the invention, has the advantages that the accuracy of edge extraction is improved, the influence on the noise is reasonably reduced, and the preparation information is provided for subsequent processing, such as further image analysis and feature point positioning.

The invention relates to an image information processing technology, and provides an ultrahigh resolution panorama speed dome AIO (All-In-One) system, a video stitching method and a spatial registration method of the ultrahigh resolution panorama speed dome AIO system. A device comprises an engine base element, a panorama stitching video camera comprising multiple cameras, a speed dome, a controller, a shell and a support. The video stitching method comprises the steps of lookup table calculation, stitching line calculation, image fusion and the like. The spatial registration method comprises the steps of establishing a panorama stitching image coordinate system, uniformly dividing a panorama stitching image into a plurality of rectangular regions, sampling, calculating a mapping relation of sampling points and calculating a non-sampling point mapping relation by adopting a bilinear interpolation method. According to the ultrahigh resolution panorama speed dome AIO system, the video stitching method and the spatial registration method, provided by the invention, not only can panorama monitoring be realized, but also a high-resolution monitoring scene can be provided at the same time, and quick and accurate focus location, zoom out and zoom in and quick and stable browsing of an interested target can be realized through registration linkage between a panorama real-time stitching camera and the speed dome.

The invention discloses a pressure vessel weld defect identification method and device based on a neural network. A computer reads in a radiographic inspection negative film image, and whether the quality of the negative film image is qualified or not is determined according to the read-in average darkness value of the negative film image; the negative film image is enhanced according to a fuzzy theory, and de-noising processing is conducted on the enhanced negative film image according to median filtering; a weld joint region is separated from the negative film image obtained after de-noising processing is conducted according to the morphological feature, and an original weld joint gray value is retained; defect filtering is conducted on the separated negative film image according to a mean filter, and a simulated ideal weld joint is formed to serve as a virtual background, the virtual background is subtracted from the original weld joint, and the weld joint defect is obtained after enhancement and de-noising are conducted. The invention further discloses the pressure vessel weld defect identification device, radiographic inspection weld joint images can be identified effectively, identification accuracy reaches 92%, the detection cost can be reduced greatly, and efficiency is improved.

A flicker correction apparatus for correcting a flicker component of an image signal obtained by imaging an object using an imaging device is provided. The apparatus comprises an image average calculation section for calculating an average of the image signal, a flicker frequency calculation section for calculating a flicker frequency, a flicker data extraction section for extracting flicker data using the average of the image signal and the flicker frequency, a flicker determination section for determining the presence or absence of a flicker phenomenon using the flicker data, a flicker correction amount calculation section for calculating a flicker correction amount using the flicker data, and a flicker correction section for removing the flicker component of the image data using the flicker correction amount.

The method comprises: creating a peak brightness table relating to the mean image level; reading out the RGB data of one frame of video signal; calculating the mean image level of said one frame of image; according to the mean image level, searching the relative peak brightness from the peak brightness table relating to the mean image level; according to the relative peak brightness, calculating the duty ratio; inputting the duty ratio into the pulse width modulator located in the backlight controlling unit of the LCD TV; adjusting the backlight brightness of the LCD TV; making the inversed control for the image to increase the signal brightness. The invention increases the contrast of TV images.

It is difficult for a material having low resistance to electron beam irradiation to obtain an electron microscopic image having a high S / N ratio. A conventional image smoothing process can improve stability of measurement, but this process has a problem of measurement errors for absolute values, reduction of sensitivity, deterioration of quality of cubic shape information and the like. In the present invention, by performing an image averaging process without deteriorating cubic shape information of a signal waveform in consideration of dimension deviation of a measurement target pattern, measurement stability is compatible with improvement of precision and sensitivity. Accordingly, it is possible to realize measurement of pattern dimensions and shapes with high precision and control of a highly sensitive semiconductor manufacturing process using the measurement.

The invention relates to an X-Y address type solid-state image pickup device manufactured by a CMOS process, and has an object to provide an X-Y address type solid-state image pickup device in which a chip area is not increased, manufacturing costs are suppressed, and an image averaging processing can be carried out. Pixel regions Pmn are arranged in a matrix form in regions defined by horizontal selection lines RWm and vertical selection lines CLn. Each of the pixel regions Pmn includes a photodiode 10, a source follower amplifier 14 for converting an electric charge of the photodiode 10 into a voltage and amplifying it to output image data, and a horizontal selection transistor 16 for outputting the image data to a predetermined one of the vertical selection lines CLn. An amplifier / noise cancel circuit 6 has a built-in image averaging circuit for carrying out an averaging processing of the image data outputted from at least two of the plurality of the pixel regions Pmn.

The utility model discloses a method and device that determines environment luminance, aiming to provide a technical proposal of determining environment luminance with the advantages of high accuracy, easy realization and low cost. The utility model is characterized in that the even luminance of images formed by the sensitization device of the imaging equipment is obtained; the environment luminance is determined according to the proportion relation of the image even luminance and the environment luminance. The utility model can be used to more accurately determine environment luminance.

The invention provides a binaryzation method of a two dimension code image. The binaryzation method comprises the following steps: identifying the magnitude of a block area of a gray level image based on the specification of the two-dimension code image, calculating the contract ratio of each block area and judging high and low of the contract ratio, arranging the threshold of the block area that the contract ratio is low and the average gray value is lower than the whole image average gray value at 255, arranging the threshold of the block area that the contract ratio is low and the average gray value is higher than the image average grey value at 0; calculating the threshold of the block area which the contract ratio is high; smoothly processing the threshold in all block areas by calculation, and comparing the relative threshold of the block area after the smooth process with each pixel of the two-dimension gray value image, and obtaining the binaryzation image. By using the method to obtain the binaryzation image of the two-dimension code image, not only speed is fast, but also the image is distortionless, so that recognition and reading of the two-dimension code is favorable.

Techniques are disclosed for determining the location of laser spot in an image, enabling wind sensing to be performed on captured images of the laser spot. Techniques can include image averaging, background subtraction, and filtering to help ensure that the Gaussian laser spot is detected in the image. Embodiments may include defining a bounding region and altering the operation of a camera such that the camera does not provide pixel data from pixels sensors corresponding pixels of outside the bounding region in subsequent image captures. Embodiments may additionally or alternatively include extracting two stereoscopic images from a single image capture.

The invention discloses a method for determining the number of acquired pressure sensitive paint (PSP) test images. The method comprises installing and statically debugging a PSP optical measurement system; completing a first PSP test and obtaining in order a pre-test background image set, a windless image set, a windy image set, and a post-test background image set by using a manner of continuously acquiring the same number of images; preprocessing the PSP image and calculating a ratio image; performing model region image segmentation and erosion processing to obtain a model region in the image; calculating the signal-to-noise ratio of the ratio image, and the signal-to-noise ratios of all ratio images; according to the average number of the images and the signal-to-noise ratio curve, determining the average number of images when the signal-to-noise ratio of the ratio image converges as the image acquisition number of the subsequent test. The method is applicable to various test environments, and can rationally control test time and improve test efficiency while ensuring data quality.

The invention discloses a method for image fusion by using multi-scale top-hat selective transform. The method comprises five major steps of: 1, extracting an effective region in an image by defining parameters in the top-hat selective transform; 2, extracting a multi-scale bright region and a multi-scale dark region of an original image through the defined top-hat selective transform by using multi-scale structural elements; 3, acquiring a bright region and a dark region of a corresponding fused image under each scale by using the maximum value aiming at the bright regions and the dark regions of different original images extracted under each scale; 4, acquiring a bright region and a dark region for final image fusion by taking the maximum value of the bright regions or the dark regions of the corresponding fused image on all scales; and 5, after a basic image is acquired by averaging the original images, adding the acquired bright region to the basic image and subtracting the acquired dark region to fulfill the aim of image fusion respectively, thereby obviously improving contrast of the fused image and image detail. The method is applied to various image based application systems and has broad application prospect.

A motion estimation method is provided for processing successive images in an image sequence, with a motion vector being associated with each of the processed images. For a current image, motion vectors associated with images that precede the current image in the sequence are selected. Candidate motion vectors are generated from the motion vectors that are selected. A motion vector is elected from among the candidate motion vectors. Information that associates the elected motion vector with the current image is stored in memory. At least one of candidate motion vectors is an acceleration vector generated from the acceleration between first and second motion vectors averaged relative to a first and second images, with the first and second images being distinct and preceding the current image in the image sequence. A motion vector averaged relative to a given image is obtained from selected motion vectors associated with images preceding the given image.

The invention discloses a real-time image content retrieval system based on full-parallel image processors. The system comprises a full-parallel array processor, an auxiliary input / output device, a control device, a computer and image retrieval software, wherein the full-parallel processor is utilized to perform large-scale parallel operation so as to ensure that the speed of extracting and matching image features is much higher than that of usual computer-based software methods and then realize real-time image content retrieval. The invention also discloses an image feature extraction method suitable to realize parallel processing. The method is suitable for parallel processing on hardware and cannot be affected by the changes of the average brightness of images. By utilizing the invention, the image content retrieval system can realize the real-time exact matching and retrieval of a large number of image data.

The invention relates to a method for controlling the back light brightness of a display, comprising the following steps: averagely dividing each image into at least two pieces; generating a brightness control signal according to the brightness information of each image and the brightness of a current environment; and controlling the back light of each image according to the brightness control signal of each image. The invention also discloses a back light control device of the display. The invention can improve the softness of a picture and heighten the contrast of the picture and can also reduce the power dissipation of equipment.

The invention provides a non-reference definition evaluating method and a non-reference definition evaluating system for a full-slice image. The non-reference definition evaluating method comprises the steps of: adopting a pyramid multi-layer data structure for storing digital images, and segmenting the images in each layer of the pyramid multi-layer data structure into a plurality of image blocks evenly; dividing a strong edge in a gradient amplitude histogram of the image blocks by adopting an Otsu threshold segmentation method, calculating to obtain strong edge intensity according to a gradient maximum value and width of the strong edge, correcting the strong edge intensity by utilizing background complexity, and determining definition of each single image block; setting a standard value of the definition, and determining whether a current image block is a clear image or a blurred image; and finally determining whether the images in each layer are clear images or not according to proportions of clear image blocks in each layer. The non-reference definition evaluating method and the non-reference definition evaluating system avoid the image definition score differences caused by image content differences, correct the influence of the image background complexity on the image definition, and realizes effective identification of the image which is partially clear and partially blurred.

The invention provides an automatic switching method of day and night modes based on an RGB sensor. The method comprises the following steps: obtaining a gain and an exposure time of the RGB sensor, and performing block statistics on Bayer data to obtain average values of color components of the RGB; determining an exposure parameter P1 according to the obtained gain and the exposure time, and calculating image average saturation S1 based on the counted average values of color components of the RGB; if (P1)TP1)&&(S1(TS1), switching the day mode into the night mode, and meanwhile opening an infrared lamp; determining an exposure parameter P2 according to the obtained gain and the exposure time, and calculating image average saturation S2 based on the counted average values of color components of the RGB; and if (P2(TP2)&&(S2)TS2), switching the night mode into the day mode, and closing the infrared lamp. By adoption of the automatic switching method provided by the invention, the accuracy of a day-night switching threshold can be improved, the misjudgment of the day-night switching mode and the problem of switching back and forth of the infrared lamp can be reduced, the hardware cost is reduced, and the system stability and reliability are greatly improved.

The invention discloses an image contrast enhancing method. The image contrast enhancing method comprises the following steps: designing a critical visual deviation enhancing function and a gradient companding function by utilizing critical visual deviation characteristics of human eye vision, and reconstructing an enhanced image from the enhanced gradient field. The image contrast enhancing method specifically comprises the following steps: firstly, calculating an original image gradient field and obtaining a critical visual deviation threshold value and a critical visual deviation enhancing function according to the average image background brightness; performing gradient companding of different degrees according to different gradient amplitude values; finally, processing by adopting the gradient field reconstructing method and linear stretching process to obtain the output image. The image contrast enhancing method disclosed by the invention is good in noise restraining and enhancing effect, clear in detail, and meets the human eye vision characteristics better; the enhanced image of the same image is repeatedly enhanced by many times, so that the enhanced image is stable in quality. Meanwhile, for the image with relatively good quality, the image is not worsened in quality after being enhanced through the method.

The invention relates to a self-adapting detection and correction method of fixed-pattern noise based on an sCMOS camera, belonging to the field of image sensors and signal processing. The destructive problems such as defective pixel, dark signal nonuniformity, photoresponse nonuniformity and dark current which are caused by the fixed-pattern noise to an image of the sCMOS camera are solved. According to the imaging characteristics of an sCMOS sensor and the generated reason and characteristic analysis of the fixed-pattern noise, a self-adapting threshold detection algorithm formula of the fixed-pattern noise is designed, meanwhile, the average manner of multiple dark field images is utilized to reduce time domain noise, by a result obtained via detection, 3*3 median filter algorithm correction is carried out on the fixed-pattern noise one by one, and effective filling treatment is carried out on an edge. The method can be used for quickly and conveniently realizing the self-adapting detection and correction of the fixed-pattern noise of the sCMOS camera, the processing speed is fast, the destructive degree to the image can be effectively reduced, and the image quality is greatly improved, so that the method has a very good use value in the field of scientific research of cameras.