Classifications

• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
  • H04N23/20—Cameras or camera modules comprising electronic image sensors; Control thereof for generating image signals from infrared radiation only

Definitions

• One or more embodiments of the invention relate generally to thermal imaging devices and more particularly, for example, to the use of blocked infrared sensors in such devices.
• Existing thermal imagers are typically implemented with multiple infrared sensors arranged in an array to capture thermal images of a target scene.
• Individual infrared sensors generally exhibit pixel-to-pixel variations such that, if left uncorrected, infrared sensors receiving the same infrared radiation may exhibit significantly different output signals.
• calibration terms may be determined during factory testing. Unfortunately, such factory testing is often time consuming and expensive. Calibration terms may also be determined periodically through the use of a temperature controlled shutter.
• the shutter may be used to temporarily block the array of infrared sensors. By capturing images of the shutter, individual offset values may be determined for individual infrared sensors. These individual offset values may be applied to subsequently captured thermal images of a target scene to provide substantially uniform performance for the infrared sensors.
• conventional shutter implementations may be prone to mechanical failure. Such shutters may also increase the cost, weight, and complexity of thermal imagers.
• Fig. 7 illustrates a flat field correction technique in accordance with an embodiment of the disclosure.
• Fig. 8 illustrates various image processing techniques of Fig. 5 and other operations applied in an image processing pipeline in accordance with an embodiment of the disclosure.
• Fig. 12 illustrates a cross-sectional side view of an infrared sensor assembly including unshielded and shielded infrared sensors in accordance with an embodiment of the disclosure.
• Fig. 13 illustrates a flow diagram of various operations using unshielded and shielded infrared sensors in accordance with an embodiment of the disclosure.
• Fig. 1 illustrates an infrared imaging module 100 (e.g., an infrared camera or an infrared imaging device) configured to be implemented in a host device 102 in accordance with an embodiment of the disclosure.
• Infrared imaging module 100 may be implemented, for one or more embodiments, with a small form factor and in accordance with wafer level packaging techniques or other packaging techniques.
• infrared imaging module 100 may be configured to be implemented in a small portable host device 102, such as a mobile telephone, a tablet computing device, a laptop computing device, a personal digital assistant, a visible light camera, a music player, or any other appropriate mobile device.
• host device 102 may include a socket 104, a shutter 105, motion sensors 194, a processor 195, a memory 196, a display 197, and/or other components 198.
• Socket 104 may be configured to receive infrared imaging module 100 as identified by arrow 101.
• Fig. 2 illustrates infrared imaging module 100 assembled in socket 104 in accordance with an embodiment of the disclosure.
• Motion sensors 194 may be implemented by one or more accelerometers, gyroscopes, or other appropriate devices that may be used to detect movement of host device 102.
• Motion sensors 194 may be monitored by and provide information to processing module 160 or processor 195 to detect motion.
• motion sensors 194 may be implemented as part of host device 102 (as shown in Fig. 1), infrared imaging module 100, or other devices attached to or otherwise interfaced with host device 102.
• Processor 195 may be implemented as any appropriate processing device (e.g., logic device, microcontroller, processor, application specific integrated circuit (ASIC), or other device) that may be used by host device 102 to execute appropriate instructions, such as software instructions provided in memory 196.
• Display 197 may be used to display captured and/or processed infrared images and/or other images, data, and information.
• Other components 198 may be used to implement any features of host device 102 as may be desired for various applications (e.g., clocks, temperature sensors, a visible light camera, or other components).
• a machine readable medium 193 may be provided for storing non- transitory instructions for loading into memory 196 and execution by processor 195.
• infrared imaging module 100 and socket 104 may be implemented for mass production to facilitate high volume applications, such as for implementation in mobile telephones or other devices (e.g., requiring small form factors).
• the combination of infrared imaging module 100 and socket 104 may exhibit overall dimensions of approximately 8.5 mm by 8.5 mm by 5.9 mm while infrared imaging module 100 is installed in socket 104.
• Infrared sensor assembly 128 may be implemented, for example, with a cap 130 (e.g., a lid) mounted on a substrate 140.
• Infrared sensor assembly 128 may include a plurality of infrared sensors 132 (e.g., infrared detectors) implemented in an array or other fashion on substrate 140 and covered by cap 130.
• infrared sensor assembly 128 may be implemented as a focal plane array (FPA).
• FPA focal plane array
• Such a focal plane array may be implemented, for example, as a vacuum package assembly (e.g., sealed by cap 130 and substrate 140).
• infrared sensor assembly 128 may be implemented as a wafer level package (e.g., infrared sensor assembly 128 may be singulated from a set of vacuum package assemblies provided on a wafer). In one embodiment, infrared sensor assembly 128 may be implemented to operate using a power supply of approximately 2.4 volts, 2.5 volts, 2.8 volts, or similar voltages. Infrared sensors 132 may be configured to detect infrared radiation (e.g., infrared energy) from a target scene including, for example, mid wave infrared wave bands (MWIR), long wave infrared wave bands (LWIR), and/or other thermal imaging bands as may be desired in particular implementations. In one embodiment, infrared sensor assembly 128 may be provided in accordance with wafer level packaging techniques.
• infrared radiation e.g., infrared energy
• MWIR mid wave infrared wave bands
• LWIR long wave infrared wave bands
• Infrared sensors 132 may be implemented, for example, as microbolometers or other types of thermal imaging infrared sensors arranged in any desired array pattern to provide a plurality of pixels.
• infrared sensors 132 may be implemented as vanadium oxide (VOx) detectors with a 17 ⁇ pixel pitch.
• VOx vanadium oxide
• arrays of approximately 32 by 32 infrared sensors 132, approximately 64 by 64 infrared sensors 132, approximately 80 by 64 infrared sensors 132, or other array sizes may be used.
• Substrate 140 may include various circuitry including, for example, a read out integrated circuit (ROIC) with dimensions less than approximately 5.5 mm by 5.5 mm in one embodiment.
• Substrate 140 may also include bond pads 142 that may be used to contact complementary connections positioned on inside surfaces of housing 120 when infrared imaging module 100 is assembled as shown in Figs. 5A, 5B, and 5C.
• the ROIC may be implemented with low-dropout regulators (LDO) to perform voltage regulation to reduce power supply noise introduced to infrared sensor assembly 128 and thus provide an improved power supply rejection ratio (PSRR).
• LDO low-dropout regulators
• PSRR power supply rejection ratio
• Fig. 4 illustrates a block diagram of infrared sensor assembly 128 including an array of infrared sensors 132 in accordance with an embodiment of the disclosure.
• infrared sensors 132 are provided as part of a unit cell array of a ROIC 402.
• ROIC 402 includes bias generation and timing control circuitry 404, column amplifiers 405, a column multiplexer 406, a row multiplexer 408, and an output amplifier 410.
• Image frames e.g., thermal images
• image frames captured by infrared sensors 132 may be provided by output amplifier 410 to processing module 160, processor 195, and/or any other appropriate components to perform various processing techniques described herein.
• FIG. 4 any desired array configuration may be used in other embodiments.
• Further descriptions of ROICs and infrared sensors may be found in U.S. Patent No. 6,028,309 issued February 22, 2000, which is incorporated herein by reference in its entirety.
• Infrared sensor assembly 128 may capture images (e.g., image frames) and provide such images from its ROIC at various rates.
• Processing module 160 may be used to perform appropriate processing of captured infrared images and may be implemented in accordance with any appropriate architecture.
• processing module 160 may be implemented as an ASIC.
• ASIC may be configured to perform image processing with high performance and/or high efficiency.
• processing module 160 may be implemented with a general purpose central processing unit (CPU) which may be configured to execute appropriate software instructions to perform image processing, coordinate and perform image processing with various image processing blocks, coordinate interfacing between processing module 160 and host device 102, and/or other operations.
• processing module 160 may be implemented with a field programmable gate array (FPGA).
• Processing module 160 may be implemented with other types of processing and/or logic circuits in other embodiments as would be understood by one skilled in the art.
• Electrical connections 126 may be electrically connected with bond pads 142 when infrared imaging module 100 is assembled.
• electrical connections 126 may be embedded in housing 120, provided on inside surfaces of housing 120, and/or otherwise provided by housing 120. Electrical connections 126 may terminate in connections 124 protruding from the bottom surface of housing 120 as shown in Fig. 3. Connections 124 may connect with circuit board 170 when infrared imaging module 100 is assembled (e.g., housing 120 may rest atop circuit board 170 in various embodiments).
• Processing module 160 may be electrically connected with circuit board 170 through appropriate electrical connections.
• infrared sensor assembly 128 may be electrically connected with processing module 160 through, for example, conductive electrical paths provided by: bond pads 142, complementary connections on inside surfaces of housing 120, electrical connections 126 of housing 120, connections 124, and circuit board 170.
• electrical connections 126 in housing 120 may be made from any desired material (e.g., copper or any other appropriate conductive material).
• electrical connections 126 may aid in dissipating heat from infrared imaging module 100.
• Socket 104 may include a cavity 106 configured to receive infrared imaging module 100 (e.g., as shown in the assembled view of Fig. 2).
• Infrared imaging module 100 and/or socket 104 may include appropriate tabs, arms, pins, fasteners, or any other appropriate engagement members which may be used to secure infrared imaging module 100 to or within socket 104 using friction, tension, adhesion, and/or any other appropriate manner.
• Socket 104 may include engagement members 107 that may engage surfaces 109 of housing 120 when infrared imaging module 100 is inserted into a cavity 106 of socket 104. Other types of engagement members may be used in other embodiments.
• Infrared imaging module 100 may be electrically connected with socket 104 through appropriate electrical connections (e.g., contacts, pins, wires, or any other appropriate connections).
• socket 104 may include electrical connections 108 which may contact corresponding electrical connections of infrared imaging module 100 (e.g., interconnect pads, contacts, or other electrical connections on side or bottom surfaces of circuit board 170, bond pads 142 or other electrical connections on base 150, or other connections).
• Electrical connections 108 may be made from any desired material (e.g., copper or any other appropriate conductive material).
• electrical connections 108 may be mechanically biased to press against electrical connections of infrared imaging module 100 when infrared imaging module 100 is inserted into cavity 106 of socket 104.
• electrical connections 108 may at least partially secure infrared imaging module 100 in socket 104. Other types of electrical connections may be used in other embodiments.
• Socket 104 may be electrically connected with host device 102 through similar types of electrical connections.
• host device 102 may include electrical connections (e.g., soldered connections, snap-in connections, or other connections) that connect with electrical connections 108 passing through apertures 190.
• electrical connections may be made to the sides and/or bottom of socket 104.
• infrared imaging module 100 may be implemented with flip chip technology which may be used to mount components directly to circuit boards without the additional clearances typically needed for wire bond connections.
• Flip chip connections may be used, as an example, to reduce the overall size of infrared imaging module 100 for use in compact small form factor applications.
• processing module 160 may be mounted to circuit board 170 using flip chip connections.
• infrared imaging module 100 may be implemented with such flip chip configurations.
• infrared imaging module 100 and/or associated components may be implemented in accordance with various techniques (e.g., wafer level packaging techniques) as set forth in U.S. Patent Application No. 12/844,124 filed July 27, 2010, and U.S. Provisional Patent Application No.
• host device 102 may include shutter 105.
• shutter 105 may be selectively positioned over socket 104 (e.g., as identified by arrows 103) while infrared imaging module 100 is installed therein.
• shutter 105 may be used, for example, to protect infrared imaging module 100 when not in use.
• Shutter 105 may also be used as a temperature reference as part of a calibration process (e.g., a NUC process or other calibration processes) for infrared imaging module 100 as would be understood by one skilled in the art.
• shutter 105 may be made from various materials such as, for example, polymers, glass, aluminum (e.g., painted or anodized) or other materials. In various embodiments, shutter 105 may include one or more coatings to selectively filter
• shutter 105 may be fixed in place to protect infrared imaging module 100 at all times.
• shutter 105 or a portion of shutter 105 may be made from appropriate materials (e.g., polymers or infrared transmitting materials such as silicon, germanium, zinc selenide, or chalcogenide glasses) that do not substantially filter desired infrared wavelengths.
• a shutter may be implemented as part of infrared imaging module 100 (e.g., within or as part of a lens barrel or other components of infrared imaging module 100), as would be understood by one skilled in the art.
• a shutter e.g., shutter 105 or other type of external or internal shutter
• a NUC process or other type of calibration may be performed using shutterless techniques.
• a NUC process or other type of calibration using shutterless techniques may be performed in combination with shutter-based techniques.
• Fig. 5 illustrates a flow diagram of various operations to determine NUC terms in accordance with an embodiment of the disclosure.
• the operations of Fig. 5 may be performed by processing module 160 or processor 195 (both also generally referred to as a processor) operating on image frames captured by infrared sensors 132.
• the scene will be the real world environment in which host device 102 is currently located.
• shutter 105 if optionally provided may be opened to permit infrared imaging module to receive infrared radiation from the scene.
• Infrared sensors 132 may continue capturing image frames during all operations shown in Fig. 5.
• the continuously captured image frames may be used for various operations as further discussed.
• the captured image frames may be temporally filtered (e.g., in accordance with the process of block 826 further described herein with regard to Fig.
• a NUC process initiating event is detected.
• the NUC process may be initiated in response to physical movement of host device 102.
• Such movement may be detected, for example, by motion sensors 194 which may be polled by a processor.
• a user may move host device 102 in a particular manner, such as by intentionally waving host device 102 back and forth in an "erase” or "swipe" movement.
• the user may move host device 102 in accordance with a predetermined speed and direction (velocity), such as in an up and down, side to side, or other pattern to initiate the NUC process.
• the use of such movements may permit the user to intuitively operate host device 102 to simulate the "erasing" of noise in captured image frames.
• a NUC process may be initiated by host device 102 if motion exceeding a threshold value is exceeded (e.g., motion greater than expected for ordinary use). It is contemplated that any desired type of spatial translation of host device 102 may be used to initiate the NUC process.
• a threshold value e.g., motion greater than expected for ordinary use. It is contemplated that any desired type of spatial translation of host device 102 may be used to initiate the NUC process.
• a user defined amount e.g., a percentage and/or threshold value
• motion may be determined on a per pixel basis, wherein only pixels that exhibit significant changes are accumulated to provide the blurred image frame. For example, counters may be provided for each pixel and used to ensure that the same number of pixel values are accumulated for each pixel, or used to average the pixel values based on the number of pixel values actually accumulated for each pixel. Other types of image -based motion detection may be performed such as performing a Radon transform.
• motion may be detected based on data provided by motion sensors 194. In one embodiment, such motion detection may include detecting whether host device 102 is moving along a relatively straight trajectory through space.
• the motion detected by motion sensors 194 may be conditioned on host device 102 exhibiting, or not exhibiting, particular trajectories.
• both a motion detection process and motion sensors 194 may be used.
• a determination can be made as to whether or not each image frame was captured while at least a portion of the scene and host device 102 were in motion relative to each other (e.g., which may be caused by host device 102 moving relative to the scene, at least a portion of the scene moving relative to host device 102, or both). It is expected that the image frames for which motion was detected may exhibit some secondary blurring of the captured scene (e.g., blurred thermal image data associated with the scene) due to the thermal time constants of infrared sensors 132 (e.g., microbolometer thermal time constants) interacting with the scene movement.
• infrared sensors 132 e.g., microbolometer thermal time constants
• FPN e.g., caused by one or more components of infrared imaging module 100
• FPN will remain fixed over at least short periods of time and over at least limited changes in scene irradiance during motion.
• image frames captured in close proximity in time and space during motion will suffer from identical or at least very similar FPN.
• scene information may change in consecutive image frames, the FPN will stay essentially constant.
• multiple image frames captured during motion will blur the scene information, but will not blur the FPN.
• FPN will remain more clearly defined in the blurred image frame provided in block 545 than the scene information.
• 32 or more image frames are accumulated and averaged in blocks 535 and 540.
• any desired number of image frames may be used in other embodiments, but with generally decreasing correction accuracy as frame count is decreased.
• a defocus operation may be performed to intentionally defocus the image frames captured by infrared sensors 132.
• one or more actuators 199 may be used to adjust, move, or otherwise translate optical element 180, infrared sensor assembly 128, and/or other components of infrared imaging module 100 to cause infrared sensors 132 to capture a blurred (e.g., unfocused) image frame of the scene.
• Other non-actuator based techniques are also contemplated for intentionally defocusing infrared image frames such as, for example, manual (e.g., user- initiated) defocusing.
• FPN e.g., caused by one or more components of infrared imaging module 100
• FPN will remain unaffected by the defocusing operation.
• a blurred image frame of the scene will be provided (block 545) with FPN remaining more clearly defined in the blurred image than the scene information.
• the defocus-based approach has been described with regard to a single captured image frame.
• the defocus-based approach may include accumulating multiple image frames while the infrared imaging module 100 has been defocused and averaging the defocused image frames to remove the effects of temporal noise and provide a blurred image frame in block 545.
• a blurred image frame may be provided in block 545 by either the motion-based approach or the defocus-based approach. Because much of the scene information will be blurred by either motion, defocusing, or both, the blurred image frame may be effectively considered a low pass filtered version of the original captured image frames with respect to scene information.
• the blurred image frame is processed to determine updated row and column FPN terms (e.g., if row and column FPN terms have not been previously determined then the updated row and column FPN terms may be new row and column FPN terms in the first iteration of block 550).
• updated row and column FPN terms e.g., if row and column FPN terms have not been previously determined then the updated row and column FPN terms may be new row and column FPN terms in the first iteration of block 550.
• the terms row and column may be used interchangeably depending on the orientation of infrared sensors 132 and/or other components of infrared imaging module 100.
• block 550 includes determining a spatial FPN correction term for each row of the blurred image frame (e.g., each row may have its own spatial FPN correction term), and also determining a spatial FPN correction term for each column of the blurred image frame (e.g., each column may have its own spatial FPN correction term).
• Such processing may be used to reduce the spatial and slowly varying (1/f) row and column FPN inherent in thermal imagers caused by, for example, 1/f noise characteristics of amplifiers in ROIC 402 which may manifest as vertical and horizontal stripes in image frames.
• row and column FPN terms may be determined by considering differences between neighboring pixels of the blurred image frame.
• Fig. 6 illustrates differences between neighboring pixels in accordance with an embodiment of the disclosure. Specifically, in Fig. 6 a pixel 610 is compared to its 8 nearest horizontal neighbors: d0-d3 on one side and d4-d7 on the other side. Differences between the neighbor pixels can be averaged to obtain an estimate of the offset error of the illustrated group of pixels. An offset error may be calculated for each pixel in a row or column and the average result may be used to correct the entire row or column. To prevent real scene data from being interpreted as noise, upper and lower threshold values may be used (thPix and -thPix). Pixel values falling outside these threshold values (pixels dl and d4 in this example) are not used to obtain the offset error. In addition, the maximum amount of row and column FPN correction may be limited by these threshold values.
• the updated row and column FPN terms determined in block 550 are stored (block 552) and applied (block 555) to the blurred image frame provided in block 545. After these terms are applied, some of the spatial row and column FPN in the blurred image frame may be reduced. However, because such terms are applied generally to rows and columns, additional FPN may remain such as spatially uncorrelated FPN associated with pixel to pixel drift or other causes. Neighborhoods of spatially correlated FPN may also remain which may not be directly associated with individual rows and columns.
• local contrast values e.g., edges or absolute values of gradients between adjacent or small groups of pixels
• scene information in the blurred image frame includes contrasting areas that have not been significantly blurred (e.g., high contrast edges in the original scene data)
• contrasting areas e.g., high contrast edges in the original scene data
• local contrast values in the blurred image frame may be calculated, or any other desired type of edge detection process may be applied to identify certain pixels in the blurred image as being part of an area of local contrast. Pixels that are marked in this manner may be considered as containing excessive high spatial frequency scene information that would be interpreted as FPN (e.g., such regions may correspond to portions of the scene that have not been sufficiently blurred). As such, these pixels may be excluded from being used in the further determination of NUC terms.
• contrast detection processing may rely on a threshold that is higher than the expected contrast value associated with FPN (e.g., pixels exhibiting a contrast value higher than the threshold may be considered to be scene information, and those lower than the threshold may be considered to be exhibiting FPN).
• the contrast determination of block 560 may be performed on the blurred image frame after row and column FPN terms have been applied to the blurred image frame (e.g., as shown in Fig. 5). In another embodiment, block 560 may be performed prior to block 550 to determine contrast before row and column FPN terms are determined (e.g., to prevent scene based contrast from contributing to the determination of such terms).
• a flat field correction process is performed on the high pass filtered blurred image frame to determine updated NUC terms (e.g., if a NUC process has not previously been performed then the updated NUC terms may be new NUC terms in the first iteration of block 570).
• Fig. 7 illustrates a flat field correction technique 700 in accordance with an embodiment of the disclosure.
• a NUC term may be determined for each pixel 710 of the blurred image frame using the values of its neighboring pixels 712 to 726. For each pixel 710, several gradients may be determined based on the absolute difference between the values of various adjacent pixels.
• a weight value may be determined for pixel 710 that is inversely proportional to the summed gradient. This process may be performed for all pixels 710 of the blurred image frame until a weight value is provided for each pixel 710. For areas with low gradients (e.g., areas that are blurry or have low contrast), the weight value will be close to one. Conversely, for areas with high gradients, the weight value will be zero or close to zero. The update to the NUC term as estimated by the high pass filter is multiplied with the weight value.
• the risk of introducing scene information into the NUC terms can be further reduced by applying some amount of temporal damping to the NUC term determination process.
• the row and column FPN filters may in general use more data to calculate the per row and per column offset coefficients (e.g., row and column FPN terms) and may thus provide a more robust alternative for reducing spatially correlated FPN than the NUC terms which are based on high pass filtering to capture spatially uncorrelated noise.
• additional high pass filtering and further determinations of updated NUC terms may be optionally performed to remove spatially correlated FPN with lower spatial frequency than previously removed by row and column FPN terms.
• some variability in infrared sensors 132 or other components of infrared imaging module 100 may result in spatially correlated FPN noise that cannot be easily modeled as row or column noise.
• Such spatially correlated FPN may include, for example, window defects on a sensor package or a cluster of infrared sensors 132 that respond differently to irradiance than neighboring infrared sensors 132.
• such spatially correlated FPN may be mitigated with an offset correction.
• the noise may also be detectable in the blurred image frame. Since this type of noise may affect a neighborhood of pixels, a high pass filter with a small kernel may not detect the FPN in the neighborhood (e.g., all values used in high pass filter may be taken from the neighborhood of affected pixels and thus may be affected by the same offset error). For example, if the high pass filtering of block 565 is performed with a small kernel (e.g., considering only immediately adjacent pixels that fall within a neighborhood of pixels affected by spatially correlated FPN), then broadly distributed spatially correlated FPN may not be detected.
• a small kernel e.g., considering only immediately adjacent pixels that fall within a neighborhood of pixels affected by spatially correlated FPN
• Fig. 11 illustrates spatially correlated FPN in a neighborhood of pixels in accordance with an embodiment of the disclosure.
• a neighborhood of pixels 1110 may exhibit spatially correlated FPN that is not precisely correlated to individual rows and columns and is distributed over a neighborhood of several pixels (e.g., a neighborhood of approximately 4 by 4 pixels in this example).
• Sample image frame 1100 also includes a set of pixels 1120 exhibiting substantially uniform response that are not used in filtering calculations, and a set of pixels 1130 that are used to estimate a low pass value for the neighborhood of pixels 1110.
• pixels 1130 may be a number of pixels divisible by two in order to facilitate efficient hardware or software calculations.
• additional high pass filtering and further determinations of updated NUC terms may be optionally performed to remove spatially correlated FPN such as exhibited by pixels 1110.
• the updated NUC terms determined in block 570 are applied to the blurred image frame.
• the blurred image frame will have been initially corrected for spatially correlated FPN (e.g., by application of the updated row and column FPN terms in block 555), and also initially corrected for spatially uncorrected FPN (e.g., by application of the updated NUC terms applied in block 571).
• a further high pass filter is applied with a larger kernel than was used in block 565, and further updated NUC terms may be determined in block 573.
• the high pass filter applied in block 572 may include data from a sufficiently large enough neighborhood of pixels such that differences can be determined between unaffected pixels (e.g., pixels 1120) and affected pixels (e.g., pixels 1110).
• a low pass filter with a large kernel can be used (e.g., an N by N kernel that is much greater than 3 by 3 pixels) and the results may be subtracted to perform appropriate high pass filtering.
• a sparse kernel may be used such that only a small number of neighboring pixels inside an N by N neighborhood are used.
• the temporal damping factor ⁇ may be set close to 1 for updated NUC terms determined in block 573.
• blocks 571-573 may be repeated (e.g., cascaded) to iteratively perform high pass filtering with increasing kernel sizes to provide further updated NUC terms further correct for spatially correlated FPN of desired neighborhood sizes.
• the decision to perform such iterations may be determined by whether spatially correlated FPN has actually been removed by the updated NUC terms of the previous performance of blocks 571-573.
• thresholding criteria may be applied to individual pixels to determine which pixels receive updated NUC terms.
• the threshold values may correspond to differences between the newly calculated NUC terms and previously calculated NUC terms.
• the threshold values may be independent of previously calculated NUC terms. Other tests may be applied (e.g., spatial correlation tests) to determine whether the NUC terms should be applied.
• Image frames captured by infrared sensors 132 may be provided to a frame averager 804 that integrates multiple image frames to provide image frames 802 with an improved signal to noise ratio.
• Frame averager 804 may be effectively provided by infrared sensors 132, ROIC 402, and other components of infrared sensor assembly 128 that are implemented to support high image capture rates.
• infrared sensor assembly 128 may capture infrared image frames at a frame rate of 240 Hz (e.g., 240 images per second).
• such a high frame rate may be implemented, for example, by operating infrared sensor assembly 128 at relatively low voltages (e.g., compatible with mobile telephone voltages) and by using a relatively small array of infrared sensors 132 (e.g., an array of 64 by 64 infrared sensors in one embodiment).
• relatively low voltages e.g., compatible with mobile telephone voltages
• infrared sensors 132 e.g., an array of 64 by 64 infrared sensors in one embodiment
• such infrared image frames may be provided from infrared sensor assembly 128 to processing module 160 at a high frame rate (e.g., 240 Hz or other frame rates).
• infrared sensor assembly 128 may integrate over longer time periods, or multiple time periods, to provide integrated (e.g., averaged) infrared image frames to processing module 160 at a lower frame rate (e.g., 30 Hz, 9 Hz, or other frame rates). Further information regarding implementations that may be used to provide high image capture rates may be found in U.S. Provisional Patent Application No. 61/495,879 previously referenced herein.
• Image frames 802 proceed through pipeline 800 where they are adjusted by various terms, temporally filtered, used to determine the various adjustment terms, and gain compensated.
• factory gain terms 812 and factory offset terms 816 are applied to image frames 802 to compensate for gain and offset differences, respectively, between the various infrared sensors 132 and/or other components of infrared imaging module 100 determined during manufacturing and testing.
• NUC terms 817 are applied to image frames 802 to correct for FPN as discussed. In one embodiment, if NUC terms 817 have not yet been determined (e.g., before a NUC process has been initiated), then block 580 may not be performed or initialization values may be used for NUC terms 817 that result in no alteration to the image data (e.g., offsets for every pixel would be equal to zero).
• block 580 may not be performed or initialization values may be used for NUC terms 817 that result in no alteration to the image data (e.g., offsets for every pixel would be equal to zero).
• column FPN terms 820 and row FPN terms 824 are applied to image frames 802. Column FPN terms 820 and row FPN terms 824 may be determined in accordance with block 550 as discussed.
• columns 818 and 822 may not be performed or initialization values may be used for the column FPN terms 820 and row FPN terms 824 that result in no alteration to the image data (e.g., offsets for every pixel would be equal to zero).
• temporal filtering is performed on image frames 802 in accordance with a temporal noise reduction (TNR) process.
• Fig. 9 illustrates a TNR process in accordance with an embodiment of the disclosure.
• Image frames 802a and 802b include local neighborhoods of pixels 803a and 803b centered around pixels 805a and 805b, respectively. Neighborhoods 803a and 803b correspond to the same locations within image frames 802a and 802b and are subsets of the total pixels in image frames 802a and 802b.
• neighborhoods 803a and 803b include areas of 5 by 5 pixels. Other neighborhood sizes may be used in other embodiments.
• Averaged delta value 805c may be used to determine weight values in block 807 to be applied to pixels 805a and 805b of image frames 802a and 802b.
• the weight values determined in block 807 may be inversely proportional to averaged delta value 805c such that weight values drop rapidly towards zero when there are large differences between neighborhoods 803a and 803b.
• large differences between neighborhoods 803a and 803b may indicate that changes have occurred within the scene (e.g., due to motion) and pixels 802a and 802b may be appropriately weighted, in one embodiment, to avoid introducing blur across frame-to- frame scene changes.
• Other associations between weight values and averaged delta value 805c may be used in various embodiments.
• the weight values determined in block 807 may be applied to pixels 805a and 805b to determine a value for corresponding pixel 805e of image frame 802e (block 811).
• pixel 805e may have a value that is a weighted average (or other combination) of pixels 805a and 805b, depending on averaged delta value 805c and the weight values determined in block 807.
• image frame 802a has been described as a presently received image frame and image frame 802b has been described as a previously temporally filtered image frame.
• image frames 802a and 802b may be first and second image frames captured by infrared imaging module 100 that have not been temporally filtered.
• Fig. 10 illustrates further implementation details in relation to the TNR process of block 826. As shown in Fig. 10, image frames 802a and 802b may be read into line buffers 1010a and 1010b, respectively, and image frame 802b (e.g., the previous image frame) may be stored in a frame buffer 1020 before being read into line buffer 1010b.
• image frame 802b e.g., the previous image frame
• line buffers lOlOa-b and frame buffer 1020 may be implemented by a block of random access memory (RAM) provided by any appropriate component of infrared imaging module 100 and/or host device 102.
• RAM random access memory
• Fig. 8 further illustrates various operations that may be performed to determine row and column FPN terms and NUC terms as discussed.
• these operations may use image frames 802e as shown in Fig. 8. Because image frames 802e have already been temporally filtered, at least some temporal noise may be removed and thus will not inadvertently affect the determination of row and column FPN terms 824 and 820 and NUC terms 817. In another embodiment, non-temporally filtered image frames 802 may be used.
• a NUC process may be selectively initiated and performed in response to various NUC process initiating events and based on various criteria or conditions.
• the NUC process may be performed in accordance with a motion-based approach (blocks 525, 535, and 540) or a defocus-based approach (block 530) to provide a blurred image frame (block 545).
• Fig. 8 further illustrates various additional blocks 550, 552, 555, 560, 565, 570, 571, 572, 573, and 575 previously discussed with regard to Fig. 5.
• row and column FPN terms 824 and 820 and NUC terms 817 may be determined and applied in an iterative fashion such that updated terms are determined using image frames 802 to which previous terms have already been applied. As a result, the overall process of Fig. 8 may repeatedly update and apply such terms to continuously reduce the noise in image frames 830 to be used by host device 102.
• FIG. 10 further implementation details are illustrated for various blocks of Figs. 5 and 8 in relation to pipeline 800. For example, blocks 525, 535, and 540 are shown as operating at the normal frame rate of image frames 802 received by pipeline 800. In the embodiment shown in Fig.
• the determination made in block 525 is represented as a decision diamond used to determine whether a given image frame 802 has sufficiently changed such that it may be considered an image frame that will enhance the blur if added to other image frames and is therefore accumulated (block 535 is represented by an arrow in this embodiment) and averaged (block 540).
• Fig. 10 also illustrates further implementation details in relation to the NUC determination process of block 570.
• the blurred image frame may be read to a line buffer 1030 (e.g., implemented by a block of RAM provided by any appropriate component of infrared imaging module 100 and/or host device 102).
• the flat field correction technique 700 of Fig. 7 may be performed on the blurred image frame.
• the rate at which row and column FPN terms and/or NUC terms are updated can be inversely proportional to the estimated amount of blur in the blurred image frame and/or inversely proportional to the magnitude of local contrast values (e.g., determined in block 560).
• the described techniques may provide advantages over conventional shutter-based noise correction techniques.
• a shutter e.g., such as shutter 105
• Power and maximum voltage supplied to, or generated by, infrared imaging module 100 may also be reduced if a shutter does not need to be mechanically operated. Reliability will be improved by removing the shutter as a potential point of failure.
• a shutterless process also eliminates potential image interruption caused by the temporary blockage of the imaged scene by a shutter.
• an absolute (e.g., not relative) radiometric value may be determined for each pixel of a thermal image of a scene captured by infrared imaging module 100.
• one or more infrared sensors 132 of infrared sensor assembly 128 may be shielded (e.g., blinded, blocked, and/or obscured) from the scene, while other infrared sensors 132 of infrared sensor assembly 128 remain unshielded and are used to capture thermal images of the scene.
• the shielded infrared sensors 132 may be used to determine an average thermographic offset reference which is used to determine an absolute radiometric value for each pixel of a thermal image captured by unshielded infrared sensors 132.
• the absolute radiometric values may be used, for example, to determine an absolute temperature for each pixel of the scene without requiring transmissions from infrared imaging module 100 toward the scene, and without requiring local temperature measurements within the scene (e.g., by sensors positioned within the scene and/or remote from infrared imaging module 100).
• the absolute radiometric values may be used in other processes as may be desired for particular applications.
• Fig. 12 illustrates a cross- sectional side view of infrared sensor assembly 128 including an unshielded infrared sensor 132A and a shielded infrared sensor 132B in accordance with an embodiment of the disclosure.
• infrared sensor assembly 128 may be implemented as a wafer level package.
• infrared sensors 132A-B may be provided as part of the wafer level package and positioned under cap 130.
• Infrared sensors 132A-B may be implemented, for example, as microbolometers or other types of thermal imaging infrared sensors arranged in any desired array pattern or other desired distribution.
• Infrared sensors 132A-B may also be thermally decoupled from substrate 140 and may be implemented with substantially the same responsivity and other operational characteristics as all infrared sensors 132 of infrared sensor assembly 128.
• Unshielded infrared sensor 132A (e.g., also referred to as an active infrared sensor) is positioned to receive infrared radiation 1210 attributable to a scene 1201 and may be used with one or more additional unshielded infrared sensors 132A to capture thermal images of scene 1201.
• cap 130 may include a top surface 1202 with an aperture 1204 therein to pass infrared radiation 1210 received through optical element 180.
• Shielded infrared sensor 132B (e.g., also referred to as a blind or blocked infrared sensor) is positioned such that it is substantially and/or completely prevented from receiving infrared radiation 1210.
• shielded infrared sensor 132B is positioned under a shield 1206 which blocks infrared radiation 1210.
• An additional shield 1208 may be provided to block additional infrared sensors if desired.
• shields 1206 and 1208 are illustrated as structures connected to sidewalls of cap 130, other types of shields may be used.
• cap 130 may be configured to block infrared radiation 1210 from shielded infrared sensor 132B based on the physical shape or configuration of cap 130, getter and/or other material provided on inside and/or outside surfaces of cap 130, and/or by other techniques.
• cap 130 may be configured in accordance with any of the techniques identified in U.S. Provisional Patent Application No. 61/469,651 filed March 30, 2011 which is hereby incorporated by reference in its entirety.
• Other shields in the form of various structures are also contemplated that completely block or partially block shielded infrared sensor 132B from various sources of infrared radiation 1210.
• infrared sensors 132A-B may be implemented, for example, in accordance with the various infrared sensors 132 of Fig. 4.
• one or more columns or rows of infrared sensors 132 shown in Fig. 4 may be implemented as shielded infrared sensors 132B, and the remainder implemented as unshielded infrared sensors 132A.
• Unshielded infrared sensor 132A may receive infrared radiation from various sources. For example, unshielded infrared sensor 132A may receive infrared radiation 1210 attributable to scene 1201 through an angle ⁇ ⁇ . Unshielded infrared sensor 132A may also receive infrared radiation 1211 attributable to optical element 180. In Fig. 12, infrared radiation 1211 is illustrated as being coincident with infrared radiation 1210 within infrared sensor assembly 128 and received through an angle ⁇ 2 coincident with angle ⁇ ⁇ .
• Unshielded infrared sensor 132A may also receive infrared radiation 1213 attributable to cap 130 through an angle ⁇ 3 .
• shield 1206 may substantially block shielded infrared sensor 132B from receiving infrared radiation 1210 and 1211 attributable to scene 1201 and optical element 180, respectively, shielded infrared sensor 132B may receive infrared radiation from various other sources.
• shielded infrared sensor 132B may receive infrared radiation 1214 attributable to shield 1206 through an angle i.
• Shielded infrared sensor 132B may also receive infrared radiation 1212 attributable to cap 130 through an angle a 2 .
• infrared sensors 132A-B Although various types of infrared radiation have been described as being received by infrared sensors 132A-B, the particular types, amounts, and angles of such infrared radiation may vary depending on the particular physical placement, configuration, and implementation of infrared sensors 132A-B, infrared sensor assembly 128, and/or other factors. Accordingly, it should be understood that the particular infrared radiation identified in Fig. 12 has been described for purposes of example, and that different combinations of infrared radiation may contribute to the actual infrared radiation detected by infrared sensors 132A-B in various embodiments.
• Infrared sensor assembly 128 may be implemented with one or more temperature sensors 1220, 1222, 1224, and 1226 which may be used to detect the temperature of optical element 180 (denoted T tica[ ), cap 130 (denoted T ), shield 1206 (denoted T shie[d ), and shield
• Such temperature sensors may be implemented, for example, as thermistors and/or other appropriate devices. Using the temperature detected for one or more such components, appropriate processing may be performed (e.g., by processor 195 in one embodiment) to determine the contribution of the various types of infrared radiation from such components (e.g., a radiometric value may be determined for each such component based on the detected temperature). It should be understood that the number, type, and placement of the various temperature sensors shown in Fig. 12 are illustrated for purposes of example, and that other configurations are contemplated. For example, in one embodiment, a temperature sensor for an entire array of infrared sensors 132 may be used to determine a temperature of shields 1206 and 1206 when positioned in proximity to the array. Where appropriate, temperatures for various portions of infrared sensor assembly 128 may be extrapolated from one or more temperature sensors provided with substrate 140 or elsewhere.
• shielded infrared sensor 132B may be used to determine an average thermographic offset reference for all infrared sensors 132 of infrared sensor assembly 128 which may be used to determine absolute radiometric values for each pixel of scene 1201 as imaged by particular unshielded infrared sensors 132A.
• the raw signal S R from one of infrared sensors 132 may be expressed as:
• R is the responsivity of infrared sensor 132
• W is the total incident radiation received by infrared sensor 132
• O is a total thermographic offset reference for infrared sensor 132.
• infrared sensors 132 may still exist between individual infrared sensors 132.
• infrared sensors 132 receiving the same total incident radiation W may nevertheless exhibit differences in their raw signals S R due to variations in their responsivity R as well as different total thermographic offset references O specific to each infrared sensor 132.
• equations 1 and 2 When applied to shielded infrared sensor 132B, equations 1 and 2 may be rewritten as:
• thermographic offset reference offset AO sh may be removed and the average thermographic offset reference O may be expressed as:
• shielded infrared sensor 132B may receive infrared radiation 1212 attributable to cap 130 and infrared radiation 1214 attributable to shield 1206.
• the radiometric value of infrared radiation 1214 attributable to shield 1206 (denoted W ⁇ T shield ) ) is a function of the temperature of shield 1206 and is scaled by angle a x .
• the radiometric value of infrared radiation 1212 attributable to cap 130 (denoted W(T ) ) is a function of the temperature of cap 130 and is scaled by angle a 2 .
• unshielded (e.g., active) infrared sensor 132A equations 1 and 2 may be rewritten as:
• the radiometric value of infrared radiation 1211 attributable to optical element 180 is a function of the temperature of optical element 180 and is scaled by angle ⁇ 2 .
• the radiometric value of infrared radiation 1213 attributable to cap 130 is a function of the temperature of cap 130 and is scaled by angle ⁇ 3 .
• an absolute radiometric value W(T scene ) may be provided for the portion of scene 1201 imaged by unshielded infrared sensor 132A which may be expressed as:
• the absolute radiometric value W(T scene ) may be further expressed as:
• W(T scene ) [s R a -3 ⁇ 4 3 ⁇ 4 + R sh - W sh - M a W(T optical ) + ⁇ 3 - W(T cap ))
• the absolute radiometric value W(T scene ) may be further expressed as:
• an absolute radiometric value W(T scene ) may be determined for each unshielded infrared sensor 132A in the array of infrared sensors 132. Appropriate processing may be performed on the absolute radiometric values W(T scene ) to determine an absolute temperature T scene of scene
• each unshielded infrared sensor 132A As imaged by each unshielded infrared sensor 132A.
• Specific values identified in equations 11, 12, and 13 for individual unshielded infrared sensors 132A e.g., ⁇ ⁇ , S ⁇ , R a , 3 ⁇ 4 , ⁇ 2 , and ⁇ 3 ) may be determined and applied individually for each individual unshielded infrared sensor 132A.
• Values based on the shielded infrared sensors 132B and characteristics of infrared sensor assembly 128 e.g., O , 3 ⁇ 4 3 ⁇ 4 . R SH .
• W sh , W(T shield ) , W(T cap ) , W(T optical ) , a 1 , a 2 , may be applied to all unshielded infrared sensors 132A.
• the absolute radiometric value W(T scene ) for each unshielded infrared sensor 132A may be determined based on various measured, known, and/or calculated terms identified in equation 13.
• raw signal may be determined by capturing a signal of scene 1201 by unshielded infrared sensor 132A.
• average raw signal S ⁇ h may be determined by averaging the captured signals of several shielded infrared sensors 132B. In another embodiment, the captured signal of a single shielded infrared sensor 132B may be used as representative of the averaged signals.
• Average responsivity R sh may be determined by averaging the known responsivity of several shielded infrared sensors 132B. In one embodiment, such responsivity may be determined based on a factory calibration and/or measurements of several shielded infrared sensors 132B. In another embodiment, a specified responsivity value may be used as representative of the average responsivity (e.g., a responsivity specification or a value selected within a known responsivity range for shielded infrared sensors 132B or all of infrared sensors 132).
• Angles a Y , 2 , ⁇ 1 , ⁇ 2 , and ⁇ 3 may be determined based on the physical implementation of infrared sensor assembly 128.
• Radiometric values W(T optical ) , W(T cap ) , and W(T shield ) may be determined based on temperature measurements of optical element
• the NUC term 817 determined for unshielded infrared sensor 132 A may effectively provide a total thermographic offset reference O a that is used to bring the performance of unshielded infrared sensor 132A into substantial uniformity with other infrared sensors 132.
• the total thermographic offset reference O a is the sum of the average thermographic offset reference O and the sensor- specific thermographic offset reference AO a .
• the average thermographic offset reference O may be expressed in terms of other known terms that may be determined as discussed above.
• sensor-specific thermographic offset reference AO a may be determined based on a difference between: the NUC term 817 determined for unshielded infrared sensor 132A; and the average thermographic offset reference O determined using the known terms of equation 6.
• Fig. 13 illustrates a flow diagram of various operations using unshielded and shielded infrared sensors 132A-B in accordance with an embodiment of the disclosure.
• the operations of Fig. 13 may be performed by appropriate components of infrared imaging module 100 such as, for example, infrared sensor assembly 128, processing module 160, processor 195, and/or others.
• temperatures are determined (e.g., detected) for various components radiating onto unshielded and shielded infrared sensors 132A-B.
• temperature sensors 1220, 1222, and 1224 and/or other appropriate devices may be used to detect temperature T optical of optical element 180, temperature T of cap 130, temperature
• the temperatures are converted to radiometric values associated with the various components.
• the radiometric value W(T optical ) of infrared radiation 1211 attributable to optical element 180, the radiometric value W(T ) of infrared radiation 1213 attributable to cap 130, and the radiometric value W(T shield ) of infrared radiation 1214 attributable to shield 1206 may be determined by appropriate processing of temperatures T optical , T cap , and T shield determined in block 1310.
• signals from infrared sensors 132 are captured.
• raw signals S are captured from all unshielded infrared sensors 132A, and raw signals Sf are captured from all shielded infrared sensors 132B.
• an average thermographic offset reference O is determined. In one embodiment, as identified in equation 6, this may be determined based on the average of raw signals Sf captured from shielded infrared sensors 132B in block 1330, the average R sh of the responsivity R sh of shielded infrared sensors 132B, and the total incident radiation W sh received by one of shielded infrared sensors 132B. As identified in equation 7, the total incident radiation W sh may be determined based on characteristics of infrared sensor assembly 128 (e.g., a Y and a 2 ) and radiometric values determined in block 1320 (e.g.,
• NUC terms 817 are determined for unshielded infrared sensors 132A. In one embodiment, NUC terms 817 may be determined in accordance with appropriate operations of Figs. 5 and 8.
• a sensor-specific thermographic offset reference AO a is determined for each unshielded infrared sensor 132A.
• each sensor-specific thermographic offset reference AO a may be determined based on a difference between the NUC term 817 determined in block 1350 for a particular unshielded infrared sensor 132A and the average thermographic offset reference O determined in block 1340.
• an absolute radiometric value W(T scene ) is determined for each unshielded infrared sensor 132A.
• each absolute radiometric value W(T scene ) corresponds to the scene-based radiation received by a particular one of the unshielded infrared sensors 132A for a pixel of a thermal image captured by infrared sensor assembly 128.
• the absolute radiometric values W(T scene ) may be determined based on various terms determined in other blocks of Fig. 13 and characteristics of infrared sensor assembly 128.
• additional processing may be performed using the absolute radiometric values W(T scene ) determined in block 1370.
• the absolute temperature In one embodiment, the absolute temperature
• T scerie °f eacn pixel of a thermal image of scene 1201 may be determined from the absolute radiometric value W(T scene ) of corresponding unshielded infrared sensors 132A. Other processing may be performed in other embodiments as desired.
• Fig. 13 permit absolute temperature determinations to be performed without relying on factory calibration operations and/or shutter-based techniques to determine pixel-to-pixel offset values between individual infrared sensors 132. Rather, by using shielded infrared sensors 132B, an average thermographic offset reference O may be determined and used with NUC terms 817 to compensate for pixel-to-pixel differences.

Landscapes

• Engineering & Computer Science (AREA)
• Multimedia (AREA)
• Signal Processing (AREA)
• Studio Devices (AREA)
• Photometry And Measurement Of Optical Pulse Characteristics (AREA)
• Transforming Light Signals Into Electric Signals (AREA)
• Radiation Pyrometers (AREA)

Applications Claiming Priority (3)

Publications (1)

Family

ID=46755089

Family Applications (1)

Country Status (3)

Families Citing this family (7)

Citations (3)

Family Cites Families (13)

• 2012
  • 2012-07-31 EP EP12751646.6A patent/EP2764685A1/de not_active Ceased
  • 2012-07-31 CN CN201280054228.8A patent/CN103907342B/zh active Active
  • 2012-07-31 WO PCT/US2012/049051 patent/WO2013052196A1/en active Application Filing

Patent Citations (3)

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events