Proof of concept, Resolution, Future Advances in resolution, Speed, Calibration
Proof of the Concept for the New Technology of Digital Tunable Filter
A new Tunable Filter utilizes video signals in electromagnetic spectra (visible light, IR, etc.) for identifications of gases, liquids and solids with the ability of much higher resolution and speeds to isolate, identify and quantify chemical compounds in space, not possible by the present DSP methodologies.
The preliminary work performed towards the proof of the concept of this innovation for filtering and identification of objects in video signals (Figures below), points to a powerful method in image recognition and classifications. As we will discuss later the single pixel filtering techniques in visible light and IR can lead to significant higher resolution for future NASA and DOD missions). The same exact technology is used in visible light and IR for identification and measurement of chemical compounds in space passively.
The preliminary work performed towards the proof of the concept of this innovation for filtering and identification of objects in video signals (Figures below), points to a powerful method in image recognition and classifications. As we will discuss later the single pixel filtering techniques in visible light and IR can lead to significant higher resolution for future NASA and DOD missions). The same exact technology is used in visible light and IR for identification and measurement of chemical compounds in space passively.
Picture M4.1 – Milky Way Galaxy (Courtesy of NASA)
Picture M4.2 - showing six pixels selected at coordinates 319-32 location
Picture M4.3 - Zoomed in image of Detected six pixels coordinates 319-32 location, resolution zero (Highest: details to be submitted upon request). Another single pixel was detected related to a different star or planet.
Resolution
Identification (Resolution) Theory
The newly patented hardware-based, Digital Tunable Filter (7,529,404) shown below (Figure 1), is posed to achieve NASA’s aim in exploration of outer space stars and finding an Earth-like planet much quicker and with far less cost. The operation of the Digital Tunable Filter is very similar to our daily life of watching T.V. It behaves the same way as our brain identification of targets. When we watch T.V. our eyes detects primary colors of a pixel (or group of pixels) and our brain compares the detected colors to a pre-recorded colors (in the brain), on the basis of association. The exception is that the technology does not to pre-load trillions of different shades of color. There are no computation for filtering and identification.
The newly patented hardware-based, Digital Tunable Filter (7,529,404) shown below (Figure 1), is posed to achieve NASA’s aim in exploration of outer space stars and finding an Earth-like planet much quicker and with far less cost. The operation of the Digital Tunable Filter is very similar to our daily life of watching T.V. It behaves the same way as our brain identification of targets. When we watch T.V. our eyes detects primary colors of a pixel (or group of pixels) and our brain compares the detected colors to a pre-recorded colors (in the brain), on the basis of association. The exception is that the technology does not to pre-load trillions of different shades of color. There are no computation for filtering and identification.
Differences in Resolutions Compared with Current DSP Methods
The proposed technology of pixel by pixel filtering is new thus it is hard to find literature to compare the new methodology of filtering and identification to the ongoing methods. Without resorting to a high level of technical comparisons of this Tunable Filter methodology and the present DSP practices (in which itself is a big task). The present tech Kalman filters, correlation filters and Fast Fourier Transforms (FFT) are currently used to filter and isolate a target from surrounding noise [4, 5]. They have widespread use in pattern recognition, class or cluster recognition in which variety of attributes of a moving object is used for identification and tracking. The filtering and detection is on the cluster bases [13] or motion attributes like velocity [12] or accelerations.
The proposed technology of pixel by pixel filtering is new thus it is hard to find literature to compare the new methodology of filtering and identification to the ongoing methods. Without resorting to a high level of technical comparisons of this Tunable Filter methodology and the present DSP practices (in which itself is a big task). The present tech Kalman filters, correlation filters and Fast Fourier Transforms (FFT) are currently used to filter and isolate a target from surrounding noise [4, 5]. They have widespread use in pattern recognition, class or cluster recognition in which variety of attributes of a moving object is used for identification and tracking. The filtering and detection is on the cluster bases [13] or motion attributes like velocity [12] or accelerations.
Future Advancements
The power behind differentiations of colors is based upon probability theory. When we throw a dice, the probability of getting any number (1 to 6) is 1/6. For three dices, the probability is 1/6^3 = 1/216. The same methodology applies for greater resolutions in which the number of dices is represented by number of channels (or primes) and the number of dots on a face of the dice is represented by number of bits of the A/D converter. Thus the number of shades of color of a single pixel (resolutions in visible light) is given by:
[1/256^3] ^2 = 1/256^6 = 1/281,474,976,710,656. = 2.81* 10^14.
With four primes and 9 bits of the A/D converter, the odds are 1/ 512 ^4 = 1/ 68,719,476,736. This is 4,096 times more power of resolution. For four primes and 12 bits of the A/D converter, the resolution is 1/281,474,976,710,656 or 16,777,216 times more power of resolution compared to the three primes and 8 bit A/D converter. For two or more pixels the resolution increases exponentially to much greater odds (equation 2).
Note: The above detection Figures of a star by the simulator, is based on 3 primes and 8 bit per prime.
The Achievable four primes and 12 bit per prime provides brightness resolutions of magnitudes of 10^14 better than 10^10 desired by NASA. This technology provides 10^4 or 10,000 time better differentiations in brightness. To separate planets from stars, the combinations of visible light and IR should provide the masking (filter) that is required to detect an Earth like planet surrounded by water. The IR signature of H2O is quite distinct. For a four channel of IR and 12 bits per A/D converter, the equation for resolutions is:
This provides resolutions of 1/281,474,976,710,656 * 281,474,976,710,656 = 1/ 2.8 * 10^28.
For two or more pixels the resolution increases exponentially to much greater odds given by.
This technology is posed to resolve the masking and starlight suppression problems economically in shorter time.
The power behind differentiations of colors is based upon probability theory. When we throw a dice, the probability of getting any number (1 to 6) is 1/6. For three dices, the probability is 1/6^3 = 1/216. The same methodology applies for greater resolutions in which the number of dices is represented by number of channels (or primes) and the number of dots on a face of the dice is represented by number of bits of the A/D converter. Thus the number of shades of color of a single pixel (resolutions in visible light) is given by:
- Probability of Detection of shade of color o a pixel = 1/d^p. Equation 1
- Probability of identification with ‘n’ number of pixels = [1/d^p]^n. Equation 2
[1/256^3] ^2 = 1/256^6 = 1/281,474,976,710,656. = 2.81* 10^14.
With four primes and 9 bits of the A/D converter, the odds are 1/ 512 ^4 = 1/ 68,719,476,736. This is 4,096 times more power of resolution. For four primes and 12 bits of the A/D converter, the resolution is 1/281,474,976,710,656 or 16,777,216 times more power of resolution compared to the three primes and 8 bit A/D converter. For two or more pixels the resolution increases exponentially to much greater odds (equation 2).
Note: The above detection Figures of a star by the simulator, is based on 3 primes and 8 bit per prime.
The Achievable four primes and 12 bit per prime provides brightness resolutions of magnitudes of 10^14 better than 10^10 desired by NASA. This technology provides 10^4 or 10,000 time better differentiations in brightness. To separate planets from stars, the combinations of visible light and IR should provide the masking (filter) that is required to detect an Earth like planet surrounded by water. The IR signature of H2O is quite distinct. For a four channel of IR and 12 bits per A/D converter, the equation for resolutions is:
- Probability of Detection = (1/d^p). (1/d^c). Equation 3
This provides resolutions of 1/281,474,976,710,656 * 281,474,976,710,656 = 1/ 2.8 * 10^28.
For two or more pixels the resolution increases exponentially to much greater odds given by.
- Probability of Detection = [1/d^p). (1/d^c)]^n. Equation 4
This technology is posed to resolve the masking and starlight suppression problems economically in shorter time.
Filter Bandwidth Settings
- The bandwidth can be any band of frequencies or a group of different (not adjacent) bands. It encompasses from the lowest narrowband, which is a single pixel to broadband frequencies for many pixels.
- There is no limitation to the spectral frequencies spacing. Any frequency can be detected as long as their intensity values are pre-loaded in the filter memory.
- For scientific instrumentation, unlike the human retina’s non linarites, the detection can be as linear (flat) as one might expect it to be. Any band of electromagnetic signals either continuous or disjoint is detected.
- The filtering and its linearity are dependent on the optical filters properties used to filter each prime. Improper optical filter matching will result in gaps as well as overlaps.
Speed
Speed
The proposed technology provides filtering without calculations. It sets upper and lower received intensities for each prime or IR and if all the primes are detected within the set values, it declares that a pixel is detected. This is done in digital hardware with no equation solving for detection of a target. Because of this, detection starts from the time that all the primary values are available to the filter plus 20 nanoseconds that is the propagation delays of the electronic components. Timing analysis of the proposed hardware based Tunable Filter (Figure 1) indicates filtering speeds of 80 milliseconds for a frame (or picture) of 4,000,000 pixels. This is a breakthrough processing times, to detect hundreds of targets compared to the present DSP practices.
Differences in Speed compared with present DSP Practices
The current concepts of FFT and Kalman filters extensively use multiplication and addition in either software or hardware to perform detection and thus long delays. As explained above their resolution is rather poor. Lack of resolutions in detection, most often forces the sensor platform to send raw data to a “center” for post processing and decision making. The delays due to the pre and post processing of data are a hindrance to real time immediate operations.
The results of studies of the present methodology with respect to speeds are outlined below:
The proposed technology provides filtering without calculations. It sets upper and lower received intensities for each prime or IR and if all the primes are detected within the set values, it declares that a pixel is detected. This is done in digital hardware with no equation solving for detection of a target. Because of this, detection starts from the time that all the primary values are available to the filter plus 20 nanoseconds that is the propagation delays of the electronic components. Timing analysis of the proposed hardware based Tunable Filter (Figure 1) indicates filtering speeds of 80 milliseconds for a frame (or picture) of 4,000,000 pixels. This is a breakthrough processing times, to detect hundreds of targets compared to the present DSP practices.
Differences in Speed compared with present DSP Practices
The current concepts of FFT and Kalman filters extensively use multiplication and addition in either software or hardware to perform detection and thus long delays. As explained above their resolution is rather poor. Lack of resolutions in detection, most often forces the sensor platform to send raw data to a “center” for post processing and decision making. The delays due to the pre and post processing of data are a hindrance to real time immediate operations.
The results of studies of the present methodology with respect to speeds are outlined below:
- Kalman filtering is probably the most commonly used algorithm for implementing the tracker, although recently Condensation algorithm [7] and mean shift algorithm [4] have shown to provide certain advantages especially in the presence of significant background clutter. The Kalman filter is primarily used for identification and tracking slow moving targets [11, 12, and 13]. Tracking is enhanced by the structural information perceived from the moving objects, which improves the classification [13]. Tracking is initiated every time a new moving object is determined in the scene and the features found to fulfill the predefined criteria [13].
- Kalman filter is considered to be too computationally infeasible for image super-resolution due to the size of the images and thus the state space involved [9]. It is assumed that a person changes its walking direction and walking speed according to Gaussian distributions, thereby, the translational velocity is assumed to lie between 0 and 150 cm/s.
- Variations of this filter have a widespread use in detections of moving cars in a parking lot, but they are not suited to track and count vehicles in a fast moving street or highway [12]. Tracking non-rigid targets in low-resolution images has long been realized as a region based correspondence problem, in which each target is mapped from one frame to the next according to its position, dimension, color and other contextual information. When multiple targets exist and their dimensions are not negligible in comparison with their velocities, occlusion or grouping of these targets is a routine event. This brings about uncertainty for the tracking, because the contextual information is only available for the group and individual targets cannot be identified. [10] [14].
- The proposed technology provides filtering without calculations. It sets upper and lower received intensities for each prime or IR and if all the primes are detected within the set values, it declares that a pixel is detected. This is done in digital hardware with no equation solving for detection of a target. Because of this, detection starts from the time that all the primary values are available to the filter plus 20 nanoseconds that is the propagation delays of the electronic components. Timing analysis of the proposed hardware based tunable filter (Figure 4) indicates filtering speeds of 80 milliseconds for a frame (or picture) of 4,000,000 pixels. This is a breakthrough processing times, to detect hundreds of targets compared to the present DSP practices. This feature should be one of the NASA’s objectives for robotic programs and discoveries of outer space targets.
Calibration
The new Digital Tunable Filter, Shown in US patent 8159568 by Ned M Ahdoot issued May 5, 2012) is for identifications of targets in an electromagnetic field. The operation of the Tunable Filter is very similar to our daily life of watching T.V. It behaves the same way as our eyes and brains detect targets, with identifying primary colors of a pixel (or group of pixels). There are no computation intensive calculations compared to the present technologies. The new filtering and identification capability can be used as an stand alone or it can lend itself to the some of the developed technologies of identification for better results.
|