-
公开(公告)号:US20210209719A1
公开(公告)日:2021-07-08
申请号:US16847864
申请日:2020-04-14
Applicant: TEXAS INSTRUMENTS INCORPORATED
Inventor: Mihir Narendra MODY , Niraj NANDAN , Rajat SAGAR , Shashank DABRAL , Anthony LELL , Brijesh JADAV
Abstract: A reconfigurable image processing pipeline includes an image signal processor (ISP), a control processor, and a local memory. ISP processes raw pixel data for a frame based on an image processing parameter and provides lines of processed pixel data to control processor via a first interface. For each region of interest (ROI) in the frame, ISP generates auto-exposure and auto-white balance (2A) statistics based on the lines for the ROI and writes them to the local memory via a second interface. Control processor reads 2A statistics from the local memory, determines the image processing parameter based on them, and provides the image processing parameter to ISP. ISP also generates an integer N bin histogram for control processor, which sums a portion of the N total bins and compares the summed bin count to a lighting transition threshold. The image processing parameter is further based on the comparison.
-
公开(公告)号:US20210027422A1
公开(公告)日:2021-01-28
申请号:US16911579
申请日:2020-06-25
Applicant: TEXAS INSTRUMENTS INCORPORATED
Inventor: Shashank DABRAL , Mihir Narendra MODY , Denis BEAUDOIN , Niraj NANDAN , Gang HUA
Abstract: A method of de-mosaicing pixel data from an image processor includes generating a pixel block that includes a plurality of image pixels. The method also includes determining a first image gradient between a first set of pixels of the pixel block and a second image gradient between a second set of pixels of the pixel block. The method also includes determining a first adaptive threshold value based on intensity of a third set of pixels of the pixel block. The pixels of the third set of pixels are adjacent to one another. The method also includes filtering the pixel block in a vertical, horizontal, or neutral direction based on the first and second image gradients and the first adaptive threshold value utilizing a plurality of FIR filters to generate a plurality of component images.
-
公开(公告)号:US20250054192A1
公开(公告)日:2025-02-13
申请号:US18929419
申请日:2024-10-28
Applicant: Texas Instruments Incorporated
Inventor: Gang HUA , Mihir Narendra MODY , Niraj NANDAN , Shashank DABRAL , Rajasekhar Reddy ALLU , Denis Roland BEAUDOIN
Abstract: Techniques for image processing including receiving input image data, wherein the input image data includes data associated with a clear color channel, receiving a color offset value associated with a color channel, wherein color values for the color channel are not provided in the input image data, based on the color offset value, generating intermediate estimated color values for the color channel, wherein generating the intermediate estimated color values includes: clipping color values that have a magnitude greater than the color offset value, and adjusting color values that have a magnitude less than the color offset value based on the color offset value, applying a color correction function to the intermediate estimated color values based on the color offset value to determine color corrected estimated color values, and outputting the color corrected estimated color values.
-
公开(公告)号:US20250030951A1
公开(公告)日:2025-01-23
申请号:US18909023
申请日:2024-10-08
Applicant: Texas Instruments Incorporated
Inventor: Gang HUA , Mihir Narendra MODY , Niraj NANDAN , Shashank DABRAL , Rajasekhar Reddy ALLU , Denis Roland BEAUDOIN
Abstract: A technique for image processing, comprising: receiving input image data, wherein the image data is companded into a first bit depth, wherein the image data includes incomplete color values for pixels of the image data, and wherein the image data is associated with a first color space, interpolating the image data to generate color values for the incomplete color values for pixels of the image data, expanding the image data from the first bit depth to a second bit depth, wherein the color values of the expanded image data have a linear dynamic range, and wherein the second bit depth is higher than the first bit depth, converting the color values for pixels of the expanded image data from the first color space to a second color space, and compressing the color values for pixels of the image data to a third bit depth, the third bit depth lower than the second bit depth, and wherein the compressed color values have a nonlinear dynamic range.
-
公开(公告)号:US20240427717A1
公开(公告)日:2024-12-26
申请号:US18819007
申请日:2024-08-29
Applicant: TEXAS INSTRUMENTS INCORPORATED
Inventor: Mihir Narendra MODY , Ankur ANKUR , Vivek Vilas DHANDE , Kedar Satish CHITNIS , Niraj NANDAN , Brijesh JADAV , Shyam JAGANNATHAN , Prithvi Shankar YEYYADI ANANTHA , Santhanakrishnan Narayanan NARAYANAN
Abstract: Systems and methods in which trace data is efficiently managed are provided. An example system includes a memory, a first interface, and a processing resource communicably coupled to the first interface and to the memory. The processing resource includes a buffer, and a first controller to transmit a set of data from the buffer with associated trace information for the set of data to the memory. A second controller transmits the set of data with the associated trace information from the memory to a second interface.
-
公开(公告)号:US20240320045A1
公开(公告)日:2024-09-26
申请号:US18675294
申请日:2024-05-28
Applicant: Texas Instruments Incorporated
Inventor: Mihir Narendra MODY , Kedar Satish CHITNIS , Kumar DESAPPAN , David SMITH , Pramod Kumar SWAMI , Shyam JAGANNATHAN
CPC classification number: G06F9/5016 , G06F9/5077 , G06F12/00 , G06F12/0223 , G06F2009/45583 , G06F9/50 , G06F9/5022 , G06N3/02 , G06N3/10 , G06N20/00
Abstract: Techniques for executing machine learning (ML) models including receiving an indication to run an ML model on a processing core; receiving a static memory allocation for running the ML model on the processing core; determining that a layer of the ML model uses more memory than the static memory allocated; transmitting, to a shared memory, a memory request for blocks of the shared memory; receiving an allocation of the requested blocks; running the layer of the ML model using the static memory and the range of memory addresses; and outputting results of running the layer of the ML model.
-
37.
公开(公告)号:US20230376442A1
公开(公告)日:2023-11-23
申请号:US18363035
申请日:2023-08-01
Applicant: TEXAS INSTRUMENTS INCORPORATED
Inventor: Kishon Vijay Abraham ISRAEL VIJAYPONRAJ , Sriramakrishnan GOVINDARAJAN , Mihir Narendra MODY
IPC: G06F13/42 , G06F13/40 , G06F12/1027
CPC classification number: G06F13/4247 , G06F13/4022 , G06F12/1027 , G06F2213/0038 , G06F2213/0024 , G06F2213/0026
Abstract: A system-on-chip (SoC) may be configured to enable a Multi-Chip Daisy Chain Topology using peripheral component interface express (PCIe). The SoC may include a processor, a local memory, a root complex operably connected to the processor and the local memory, and a multi-function endpoint controller. The root complex may obtain forwarding information to configure routing of transactions to one or more PCIe endpoint functions or to the local memory. The root complex may initialize, based on the forwarding information, access between a host and the one or more PCIe endpoint functions. The multi-function endpoint controller may obtain a descriptor and endpoint information to configure outbound portals for transactions to at least one remote host. The multi-function endpoint controller may establish a communication path between the host and a function out of a plurality of functions.
-
公开(公告)号:US20230291864A1
公开(公告)日:2023-09-14
申请号:US17690829
申请日:2022-03-09
Applicant: TEXAS INSTRUMENTS INCORPORATED
Inventor: Mihir Narendra MODY , Brijesh JADAV , Gang HUA , Niraj NANDAN , Rajasekhar Reddy ALLU , Ankur ANKUR , Mayank MANGLA
CPC classification number: H04N5/2628 , G06T3/40 , G06T7/12 , G06T7/60 , G06T11/00
Abstract: A technique for determining regions and block sizes for configuring a perspective transformation engine including determining a set of scale ratios for images captured by a camera, generating a scale ratio image based on the set of scale ratios, determining a set of boundary ranges for the scale ratio image, generating a binary scale ratio image using the set of scale ratios of the scale ratio image, determining a set of regions based on the set of boundary ranges for the binary scale ratio image, determining a block size for each region of the determined set of regions, and outputting the determined set of regions and the determined block sizes.
-
39.
公开(公告)号:US20230199339A1
公开(公告)日:2023-06-22
申请号:US17555145
申请日:2021-12-17
Applicant: TEXAS INSTRUMENTS INCORPORATED
Inventor: Gang HUA , Rajasekhar Reddy ALLU , Mihir Narendra MODY , Niraj NANDAN , Mayank MANGLA , Pandy KALIMUTHU
CPC classification number: H04N9/04517 , G06T1/60 , G06T3/4015 , H04N9/04551
Abstract: In the advanced driver-assistance systems (ADAS) field, RAW sensor image processing for machine vision (MV) applications can be of critical importance. Due to red/green/blue (RGB) image components being focused by the lens at different locations in image plane, the lateral chromatic aberration (LCA) phenomenon may sometimes be observed, which causes false color around edges in the final image output, especially for high contrast edges, which can impede MV applications. Disclosed herein are low-latency, efficient, optimized designs for chromatic aberration correction (CAC) modules. In some embodiments, an in-pipeline CAC design is used that: is configured to perform on-the-fly CAC without any out-of-pipeline memory traffic; enables use of wide dynamic range (WDR) sensors; uses bicubic interpolation; supports vertical and horizontal chromatic aberration red/blue color channel offsets, reduces CAC line memory requirements, and supports flexible look-up table (LUT) down-sampling factors to improve the spatial precision of correction and accommodate popular image sensor resolutions.
-
公开(公告)号:US20230171512A1
公开(公告)日:2023-06-01
申请号:US17538833
申请日:2021-11-30
Applicant: Texas Instruments Incorporated
Inventor: Gang HUA , Mihir Narendra MODY , Niraj NANDAN , Shashank DABRAL , Rajasekhar Reddy ALLU , Denis Roland BEAUDOIN
CPC classification number: H04N9/04515 , H04N9/67 , H04N9/68
Abstract: A technique for image processing, comprising: receiving input image data, wherein the image data is companded into a first bit depth, wherein the image data includes incomplete color values for pixels of the image data, and wherein the image data is associated with a first color space, interpolating the image data to generate color values for the incomplete color values for pixels of the image data, expanding the image data from the first bit depth to a second bit depth, wherein the color values of the expanded image data have a linear dynamic range, and wherein the second bit depth is higher than the first bit depth, converting the color values for pixels of the expanded image data from the first color space to a second color space, and compressing the color values for pixels of the image data to a third bit depth, the third bit depth lower than the second bit depth, and wherein the compressed color values have a nonlinear dynamic range.
-
-
-
-
-
-
-
-
-