1. A method of producing a metal matrix composite material, comprising the steps of:
(a) mixing a metal powder and a ceramic powder to prepare a mixed powder;
(b) providing a metal casing having a lower casing member and an upper closing member, the upper closing member is adapted to seal the metal casing closed;
(c) placing the mixed powder into the lower casing member;
(d) sealing the metal casing filled with the mixed powder by placing the upper closing member on the lower casing member so as to prepare a pre-rolling assembly;
(e) preheating the pre-rolling assembly in such a manner so as to maintain the mixed powder in a powder state; and
(f) rolling the pre-rolling assembly following said step of preheating to obtain the metal matrix composite material, where the metal matrix composite material includes a pair of metal plates having the mixed powder therebetween.
2. The method as defined in claim 1, wherein the step (c) further includes mechanical compacting the mixed powder in the metal casing to increase a density of the mixed powder.
3. The method as defined in claim 1, wherein the upper closing member and the lower casing member are made of the same material.
4. The method as defined in claim 1, wherein the upper closing member includes an open bottom and is adapted to be fitted onto an outer peripheral surface of the lower casing member from above in a surrounding manner so as to cover the lower casing member.
5. The method as defined in claim 1, wherein said step (f) includes positioning a longitudinal axis of the pre-rolled assembly to extend along a rolling direction and a surface thereof to be rolled is substantially disposed in a horizontal direction; and
prior to said step (f), said method further includes reinforcing an outer peripheral surface of the metal casing with a reinforcing member for said step of rolling.
6. The method as defined in claim 5, said method further includes providing the reinforcing member of a same material as that of the metal casing.
7. The method as defined in claim 5, wherein said step of reinforcing includes using a reinforcing member having a thickness which is at least 4% of a length of the metal casing along a direction orthogonal to the rolling direction.
8. A method of producing an aluminum matrix composite material, comprising the steps of:
(a) mixing an aluminum powder and a ceramic powder to prepare a mixed powder;
(b) providing an aluminum lower casing having a rectangular shape with an open top, and an aluminum closing member formed in a shape adapted to hermetically close the open top of the lower casing;
(c) placing the mixed powder into the lower casing;
(d) closing the open top of the lower casing by the closing member so as to prepare a pre-rolling assembly;
(e) preheating the pre-rolling assembly in such a manner so as to maintain the mixed powder in a powder state; and
(f) rolling the preheated assembly to obtain an aluminum matrix composite material, where the aluminum matrix composite material includes a pair of aluminum plates having the mixed powder therebetween.
9. The method as defined in claim 8, wherein the step (c) further includes mechanical compacting the mixed powder in the aluminum lower casing to increase a density of the mixed powder.
10. The method as defined in claim 8, wherein the aluminum closing member includes an open bottom and is adapted to be fitted onto an outer peripheral surface of the aluminum lower casing from above in a surrounding manner so as to cover the aluminum lower casing.
11. The method as defined in claim 8, wherein said method includes providing the aluminum closing member formed to have a slightly greater size than that of the aluminum lower casing.
12. The method as defined in claim 8, wherein said step of preheating includes preheating the pre-rolling assembly in an atmosphere to a temperature of 300 to 600\xb0 C.
13. The method as defined in claim 8, wherein said step of preheating includes preheating in an ambient atmosphere.
14. The method as defined in claim 8, wherein said step of preheating includes preheating in an inert gas atmosphere.
15. The method as defined in claim 8, wherein said step of preheating includes preheating in a vacuum atmosphere.
16. The method as defined in claim 8, wherein said step of rolling includes subjecting the pre-rolling assembly to hot rolling at a draft ranging from 10 to 70%.
17. The method as defined in claim 16, wherein said step of rolling includes subjecting the pre-rolling assembly to said hot rolling and then to warm rolling at a temperature of 200 to 300\xb0 C.
18. The method as defined in claim 17, wherein said step of rolling includes subjecting the pre-rolling assembly to a first warm rolling and then to a second warm rolling at a temperature of 200\xb0 C. or less.
19. The method as defined in claim 8, wherein said method further includes the step of heat treating at a temperature of 300 to 600\xb0 C. following said step of rolling.
20. The method as defined in claim 8, wherein:
wherein said step (f) includes positioning a longitudinal axis of the pre-rolled assembly to extend along a rolling direction and a surface thereof to be rolled is substantially disposed in a horizontal direction; and
prior to said step (f), said method further includes reinforcing at least both side outer peripheral surfaces of the aluminum closing member, which extend in the rolling direction.
21. The method as defined in claim 20, wherein said reinforcing step includes fixing first and second reinforcing members to respective ones of opposed lateral surfaces of the aluminum closing member each parallel to the rolling direction, in such a manner so as to extend along the rolling direction, and fixing third and fourth reinforcing members to respective ones of front and rear surfaces of the aluminum closing member each orthogonal to the rolling direction, in such a manner so as to extend along a direction orthogonal to the rolling direction.
22. The method as defined in claim 20, wherein said step of reinforcing includes using first and second reinforcing members, each of the first and second reinforcing members having a width which is at least 4% of a length of the aluminum closing member along a direction orthogonal to the rolling direction.
23. The method as defined in claim 8, wherein said step (f) includes rolling the pre-rolling assembly to a shape ratio which is defined as H0SQRT(R*(H0\u2212H1)), wherein: H0 is a thickness of the pre-rolling assembly; H1 is a thickness of the assembly after rolling; R is a radius of a mill roll; and SQRT(R*(H0\u2212H1) is a rolled amount of the assembly per revolution of the mill roll, and the shape ratio satisfying the following inequality:
H0SQRT(R*(H0\u2212H1))\u22661.0 or
H0SQRT(R*(H0\u2212H1))\u22672.2.
The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.
1. An apparatus for controlling automatic focus, comprising:
(a) a computer processor configured for controlling an imaging device and associated focus control element;
(b) programming executable by said computer processor for carrying out an autofocusing process comprising:
(i) capturing object images;
(ii) estimating depth between captured image pairs to arrive at a depth estimation;
(iii) performing slope correction between a blur matching curve and a calibrated model;
(iv) determining a weighted mean of depth estimations and variance which spans both past and present depth estimations; and
(v) adjusting focus with a focusing movement performed in response to said weighted mean of the depth estimations.
2. The apparatus recited in claim 1, further comprising programming executable on said computer processor for performing slope correction of blur matching data in response to multiplying matching iteration numbers by a constant to accurately fit the calibrated model of blur matching.
3. The apparatus recited in claim 1, further comprising programming executable on said computer processor for performing slope correction of a calibrated model of blur matching by multiplying slope by a constant to accurately fit matching data samples.
4. The apparatus recited in claim 1, further comprising programming executable on said computer processor for repeating steps (i) through (v) until a desired level of focus accuracy is obtained.
5. The apparatus recited in claim 4, wherein said desired level of focus accuracy is obtained when it is determined that said variance reaches a value that indicates sufficient confidence that a proper focus position has been attained.
6. The apparatus recited in claim 1, further comprising programming executable on said computer processor for: (i) terminating autofocusing at current focus position, or (ii) assigning a large variance to the estimation result and continuing to execute autofocusing, or (iii) discarding depth estimations and capturing another pair of object images from which to make another depth estimation toward continuing said autofocusing, when it is determined that an excessive number of focusing movements has been performed.
7. The apparatus recited in claim 1, wherein focus is adjusted in response to each repetition of the autofocusing process by combining present and previous depth estimations instead of only utilizing recent depth estimations.
8. The apparatus recited in claim 1, wherein said depth estimation is performed between captured image pairs, each image of which is taken at a different focus position, and estimating actual depth based on determining an amount of blur difference between captured images within the image pair.
9. The apparatus recited in claim 1, wherein said depth estimation is performed using maximum likelihood estimation (MLE) of subject depth.
10. The apparatus recited in claim 1, wherein said depth estimation is determined in response to a weighted mean d given by
d
_
=
\u2211
i
=
1
N
\u2062
(
d
i
\u03c3
i
2
)
\u2211
i
=
1
N
\u2062
(
1
\u03c3
i
2
)
in which N represents the number of lens movements during the autofocus process, di is depth estimation from ith image pair, and \u03c3i2 is variance for di.
11. The apparatus recited in claim 1, wherein said variance comprises a measure of confidence, which is generated by said programming executable on said computer processor in response to consideration of all depth estimations made during said autofocusing process.
12. The apparatus recited in claim 1:
wherein said variance comprises a measure of confidence, which is generated by said programming executable on said computer processor, in response to depth estimations which are predicted during said autofocusing process; and
wherein said variance is determined as a weighted mean \u03c3 d2 given by
\u03c3
\u2062
2
d
=
1
\u2211
i
=
1
N
\u2062
(
1
\u03c3
i
2
)
in which N represents number of unbiased depth estimations performed during said autofocusing process and \u03c3i2 is variance for the ith depth estimation.
13. The apparatus recited in claim 1, wherein said apparatus is a component of a still image camera.
14. An apparatus for electronically capturing images, comprising:
(a) an imaging device;
(b) a focus control element coupled to said imaging device and providing a focal range to said imaging device;
(c) a computer processor coupled to the imaging device and said focus control element;
(d) programming executable on said computer processor for carrying out an autofocusing process comprising:
(i) capturing object images;
(ii) estimating depth between captured image pairs to arrive at a depth estimation;
(iii) performing slope correction between a set of blur matching results and a calibrated model when a maximum absolute deviation is larger than a predetermined threshold;
(iv) determining a weighted mean of depth estimations and variance across present and past collected image pairs; and
(v) adjusting focus in response to said weighted mean of the depth estimations.
15. The apparatus recited in claim 14, further comprising programming executable on said computer processor for performing slope correction of blur matching data in response to multiplying matching iteration numbers by a constant to accurately fit the calibrated model of blur matching.
16. The apparatus recited in claim 14, further comprising programming executable on said computer processor for performing slope correction of a calibrated model of blur matching by multiplying slope by a constant to accurately fit matching data samples.
17. The apparatus recited in claim 14, further comprising programming executable on said computer processor for repeating steps (i) through (v) of claim 14 until a desired level of focus accuracy is obtained.
18. The apparatus recited in claim 17, wherein said desired level of focus accuracy is obtained when it is determined that said variance reaches a value that indicates sufficient confidence that a proper focus position has been attained.
19. The apparatus recited in claim 14, wherein said programming executable on said computer processor is configured for estimating depth between captured image pairs in response to a focus matching model which is based on blur differences determined in response to contrast changes detected as subject distance of an object whose image is being captured changes through at least a portion of the focal range of the imaging device.
20. A method of automatically adjusting camera focus, comprising:
(a) capturing a first object image by a camera device;
(b) capturing an additional object image by the camera device;
(c) estimating depth between captured image pairs in response to inputting blur difference values between captured object images into a focus matching model which is solved to generate a depth estimation;
(d) performing slope correction between a set of blur matching results and a calibrated model;
(e) determining the weighted mean of depth estimates and variance across past and present depth estimates; and
(f) adjusting camera focus position with a camera focusing movement within an autofocusing process in response to said weighted mean of the depth estimations.
21. The method recited in claim 20, further comprising:
performing slope correction of blur matching data in response to multiplying matching iteration numbers by a constant to more accurately fit the calibrated model of blur matching.
22. The method recited in claim 20, further comprising performing slope correction of a calibrated model of blur matching by multiplying slope by a constant to more accurately fit matching data samples.