JPS6250893B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a reproduction apparatus for variable area optical sound tracks, and more particularly to an improved apparatus for scanning such audio tracks.

Variable area optical sound tracks on motion picture film have been used in substantially the same way since the advent of sound motion pictures. Initially, a single monophonic optical audio track was used, the width of the white area being proportional to the amplitude of the modulated signal being recorded. Later amendments intended to reduce sound distortion proposed bilateral tracks having the same modulation information and the same pattern, and the provision of double sets of such bilateral tracks adjacent to each other. It was done. In yet another revision, it has been proposed to separately modulate the dual bilateral tracks for stereo sound reproduction.

Commercial projectors in use today still use essentially the same illumination and light sensing equipment as those of the first projection systems used to read variable area optical audio tracks. In other words, a light source and a narrow mechanical slit provide linear irradiation, and a single photocell is used for detection; in the case of a double bilateral track for stereo, two photocells are installed. There is.

An increasingly important problem in the reproduction of optical audio tracks is widening the frequency range and suppressing noise. Optical audio tracks are particularly susceptible to impact noise from dirt and scratches, which accumulate with each projection of the film. Other types of noise include particle noise within the white area of the track and noise generated within the photocell, both of which are essentially proportional to the width of the track being played and have a noise modulation effect when playing the audio signal. It gives rise to

Various techniques have been proposed to improve the quality of optical audio tracks, and even in the 1950s and 1960s, when attempts were made to widely adopt magnetic audio tracks, optical audio tracks were still better than thought. It has also been suggested that there are no defects. A useful document that discusses the history and possibilities of optical audio tracks is S.M.
Journal of
720 of Volume 84 of SMPTE, published in September 1975.
``The Production of Wide―Range Low―'' presented by Loan Allem on page 729, ``The Production of Wide―Range Low―
Distortion Optical Sound Tracks Utilizing the
There is a paper called ``Dolby Noise Reduction System''. A list of cited documents is attached to this paper.

A prior attempt to find another method of optical audio track reproduction was disclosed in U.S. Pat. No. 2,347,084. A very small scanning spot is repeatedly scanned across the width of the track and detected by a single photoresponsive cell to provide an essentially two-level pulse width modulated signal. This signal is applied to a limiter to suppress noise and integrated to obtain a voice signal with varying amplitude. This system is useful in eliminating a significant amount of the noise caused by black dots present in white areas of the track. The black areas produce virtually no noise unless the negative is imperfect and there are white spots within the black areas. However, despite the above-described scanning scheme, any black dots in the white area will result in noise.

An improved scanning system is disclosed in British patent application no.
It is described in the specification of No. 37292. This UK patent application was filed on August 11, 1975 and is no.
No. 603,671, this U.S. patent application is the parent application of U.S. Patent Application No. 778,870, which was filed on March 18, 1977 as a continuation-in-part application thereof. In this system, the knowledge that noise originally occurs in the white areas of the track is used to suppress the output signal using a squelch circuit when the transition from the black area to the white area of the track is detected during each scan. This suppresses noise. A reset is performed after each scan. A refinement of this technique involves scanning in two directions and also takes advantage of the transition from white to black areas in the bilateral track. Signal delay devices and logic circuits are used for this purpose.

A problem with these conventional scanning devices described above is that the optical scanning mechanisms are complex and require extensive modification to be adapted to film projectors currently in use. Furthermore, long-term reliability and the need for less frequent adjustment are essential requirements for commercial projection equipment. Conventional scanning devices use scanning techniques such as cathode ray tube, mechanical, laser, or electro-optical scanning, and are therefore prone to the above-mentioned problems.

The present invention provides an improved optical soundtrack playback system. This reproducing device is characterized by simultaneously processing and combining signals from sensors arranged in a linear array. Each sensor samples a small portion of the soundtrack as the motion picture film carrying the soundtrack moves vertically past the sensor.
Implementation of the invention is made more feasible by recognizing that the invention may utilize fewer sensors than may be considered necessary. In the present invention, a sensor sampling interval narrower than the width of the uncertainty transitioning from a white area to a black area of the optical soundtrack does not provide any further information, but a small number corresponding to such a relatively narrow interval It is recognized that undesirable quantization noise is introduced when using sensors such as In spite of this, in the present invention, such narrow sensor spacing and therefore a small number of sensors are set. To achieve this, the present invention provides sensor sampling intervals such that as a result of vertical movement of the soundtrack, the uncertain region of the transition boundary is randomly sampled. The uncertain part is converted into a broadband dither signal. That is, the effect of the present invention is that the correlation between the signals from the linearly arranged sensors and the quantization noise caused by a small number of sensors is reduced. than would be required if done without dithering in this way.
This makes it possible to quantize high-quality audio signals using a much smaller number of sensors. Thus, the present invention makes it practical to perform parallel sensing and parallel signal processing using a relatively small number of sensors.

The circumstances leading to the present invention are as follows. It is clear that if a group of photosensors could be arranged to detect the entire varying width of an optical audio track, the need for scanning the track with a light beam would be eliminated. It is also possible to optically magnify the track, but for practical use in currently used projectors, the sensor array should be located directly behind the film plane. However, an obstacle to using such a sensor array is the problem of quantization noise. In the audio field, approximately 10,000 or more quantization stages are required to quantize audio signals with high quality. The total width of a single stereo optical audio track on 35mm film is 0.033 inches (approximately 0.838 mm), or approximately 0.016 inches (approximately 0.419 mm) each when recorded bilaterally, so 10,000 The photo sensor element is 1.6 microinches (approximately 0.00004
This requires rows of photosensors arranged at intervals of millimeters. This is out of the question given the current state of microelectronic technology. Even if an optical magnifying device is used, it is extremely difficult to install it in a current projector, and it is highly inconceivable to provide a group of 40,000 photosensors over the entire width of a stereo track.

By studying the above-mentioned problem, the following promising results were obtained. It is known to reduce the correlation between the input signal and quantization noise by adding a wideband dither signal to the input signal when quantizing an audio signal. This significantly reduces disturbances due to quantization noise. If the dither signal is sufficiently large, a small number of quantization stages is sufficient. When a variable area optical audio track is observed under a microscope,
The boundary between the white and black areas of the track is not sharp, but rather has an uneven edge due to the presence of an area of uncertainty due to film grain. This region of uncertainty extends over a distance of 0.0001 inch to 0.0002 inch. All information transmitted in a variable area soundtrack is done by transitioning from "white" to "black," but the presence of this region of uncertainty limits the accuracy or resolution of the information. . Therefore, if the sampling interval is less than the uncertainty width, no further information can be obtained due to the uncertainty region. However, sampling performed with a width slightly smaller than the uncertainty width (sampling 2 to 3 times for the uncertainty region) is extremely effective. This is because such sampling results in the uncertain region being randomly sampled and converted into a dither signal. Thus, in the present invention, the uncertainty region is used as a unique dither signal, and by adding this dither signal to the input signal, the soundtrack is quantized by, for example, a quantization stage of approximately 0.0025 mm (0.0001 inch). can. Therefore, each half of a bilaterally recorded track can be quantized with only 160 quantization stages, which is much smaller than the 10,000 required to conventionally quantize an audio signal. .

Set the pitch of the photo sensors in the photo sensor row to 0.0001
If measured in inches (approximately 0.0025 millimeters), the integration density would be about 10 times that of what is currently done in microelectronics. Although new masking and etching methods will be required, this increase in integration density does not seem entirely out of the question. In any event, optical magnification methods can be used until the required microelectronic fabrication methods become available. Therefore, the key point of the present invention is the discovery and recognition that optical audio tracks have inherent properties that make it possible to implement quantized playback devices.

In some embodiments of the invention, a clock circuit and a switch sequentially scan the photosensor elements at high speed. The pulse width modulated output signal is then limited or otherwise processed as in conventional scanning devices. Although the embodiments described above are feasible, it is desirable to simplify the device as much as possible. Therefore, another point of the invention is the recognition that the provision of a digital readout device eliminates the need for lateral scanning of the audio track, and eliminates the need for clock circuit switching circuits, which are quite complex in structure, and pulse width modulated signal demodulation circuits. It is based on According to the invention based on this recognition, the outputs of each photosensor are processed in parallel. Each sensor output is routed directly to its own amplification and limiter circuit. Most spurious outputs due to dirt or scratches are removed by the action of the limiter. Noise can be further reduced by adding the output of the limiter circuit to the logic circuit. Then, the outputs of all the circuits are combined and smoothed to obtain an audio output signal. All of the required circuit structures are very simple, so that the photosensor elements, amplifiers, limiters, noise suppression logic, synthesis circuits and audio output amplifiers can all be integrated on a single integrated circuit chip. According to the foregoing numbers, 640 sensing and processing channels would be required, whereas in a commercial projector application the final output of the integrated circuit would need to include only two audio channels.

By selecting the photodiode portions to be treated as separate groups, the present invention allows all variable Applicable to areal optical audio tracks. Two, three,
By using four or more tracks and a corresponding processing system, it is possible to generate surrounding sound effect signals such as left side signal, center signal, right side signal, rear side signal, etc., as well as control signals for special purposes. can be generated.

Since the apparatus of the present invention uses a conventional projector illumination source, the only necessary modifications to the projector's mechanical structure are the use of photodiodes and processing ICs in place of photocells. It may be possible, and sometimes desirable, to provide suitable optical arrangements to effectively place the photodiode at the imaging surface of the film track.

The above and other objects and features of the invention will become apparent from the following description of preferred embodiments with reference to the drawings.

FIG. 1 illustrates the relationship between photographic density and distance when scanning across a variable area bilateral audio track. Pulse width W in Figure 1
Waveform information is stored within. The initial black level is essentially a constant level so there is no noise. However, since the white areas of the track usually contain black particles, the amplitude of the white level varies substantially. Unless such level fluctuations are removed by a limiter, the reproduced audio signal will contain considerable noise.
However, even this amplitude limitation does not eliminate all amplitude fluctuations, and there is also an area of uncertainty at the transition from the black level to the white level (and from the white level to the black level) as shown in Figure 2. Noise generated by the noise is also not removed. FIG. 2 is a partially enlarged view of FIG. 1 to show the rise pattern of the signal at the transition portion. The dotted line indicates a rise that may otherwise occur. For this reason,
Depending on the amplitude limit level, the apparent transition position from the black level to the white level changes, causing noise in the reproduced signal.

FIG. 3 shows the general arrangement of the apparatus of the present invention, which comprises a conventional lamp 10 and a machine for horizontally illuminating a portion of a narrow variable area audio track 14 on a motion picture film 16. It has a target slip 12. Although bilateral audio tracks are shown, the invention is also applicable to films having multiple tracks. The light modulated by the audio track 14 is detected by a light receiving device.
This light receiving device consists of a group of solid-state sensors 1 each detecting a minute portion of the width of the audio track.
It consists of 8 columns. If necessary, various lenses may be provided between the lamp 10, the slit 12, the film 16, and the sensor 18 so that the image of the slit is optically formed on the film surface and the sensor is placed on the track image forming surface. can. Processing circuit 20 receives the output of sensor 18 and generates a reproduced audio signal.
One object of the present invention is to provide a device that includes a sensor and a processing circuit on a single integrated circuit chip.

A number of sensors 18 are placed so as to obtain one sample every approximately 0.0025 mm (0.0001 inch) across the soundtrack. That is, approximately 640 sensors are provided, corresponding to an effective working width of approximately 1.6 mm (0.064 inch), excluding the stereo guard band, from a standard stereo optical motion picture film track having a width of approximately 1.9 mm (0.076 inch). This number is large enough to remove quantization noise, but not so large that it oversamples parts of the uncertainty region. In other words, irregular sampling is performed such that about 2 to 3 sampling outputs are obtained for each uncertainty region, and the resulting psychoacoustically acceptable noise is replaced by the unpleasant quantization noise. It is intended to be replaced.

In the preferred embodiment shown in FIG.
Each sensor 24 forming the
6b..., and the output of the limiter is applied to the combiner 48. In its simplest form, the synthesizer 48 sums the applied signals and applies the result to an integrator or audioband filter 40 to obtain a reproduced audio signal.

When the sensors 24 are not uniformly spaced, e.g., when the sensors 24 are arranged such that the density of the sensors 24 is higher in the center of the track area than in the other areas, the output signals are appropriately adjusted in the combiner 48. Weighting is used to compensate for density differences.

In the case of multiple audio tracks such as bilateral audio tracks for stereo, the sensor 2 shown in FIG.
4 is divided into corresponding groups so that each group of sensors reads a corresponding track, and these sensor groups are connected to separate processing circuits to reproduce separate audio signals from each track. These signals are then applied to separate combiners.

Although the action of the limiters 36 substantially reduces noise due to dirt scratches, further noise reduction can be achieved by connecting a logic circuit to the output of each limiter. For example, in FIG. 4, a logic circuit 37b is provided which receives the output of the limiter 36b.
A similar logic circuit (not shown) is connected to another limiter 37.
It can be connected to outputs such as a and 37c. Scratch noise is evident by uneven output of the limiter. For example, in the white area of the track the polarity of the limiter output is all positive. Deep scratches cause the output of one or a few limiters to go negative in polarity. The presence of an abnormal signal by comparing the polarity of the output of one limiter, e.g., limiter 36b, with the polarity of the output of the limiters on either side of it, e.g., limiters 36a, 36c, or a limiter two or more points away. can be detected. If the polarity of the output of a particular limiter is different from the polarity of the output of its reference limiter, the former output is invalidated and the latter, ie, the output of the reference limiter, is used in its place.

An example of a logic circuit for determining whether the limiter output is spurious is a combination of an OR gate 44, an AND gate 46, an AND gate 50, and an OR gate 52 as shown in FIG. That is, the output M of a certain limiter is compared with, for example, the outputs M-1 and M+1 of the limiters on both sides thereof.
If the polarities are the same, the output signal is correct, but
When the output polarity is different, the output polarity of the reference sensor is output. The logic circuitry can be simplified by removing logic elements 50, 52, but in this case the abnormal white signal produced in the black area cannot be changed to a black signal.

In practice, amplifier 34, limiter 36, and logic circuit 37 may be combined into a single electronic circuit that provides the logic input to combiner 48, with these circuits suitably connected.

FIG. 6 shows another embodiment of the invention in which processing circuitry 20 includes means for electronically scanning an array of photosensors 18. The scanning device is functionally illustrated as a scanning controller 30, which controls switches 32 selectively connected to each sensor 24. An appropriate scanning rate is about 100 kilohertz, so that it can support audio frequencies up to about 20 kilohertz.

The output of switch 32 is applied to an amplifier 34, which is preferably a low noise wideband amplifier. Preferably, the amplifier further converts the current output of the photosensor into a voltage.
The output of the amplifier is amplitude limited, ie, level clipped, by limiter 36 to remove most of the fluctuations contained in the signal from the black/white areas and to apply a substantially two level signal to logic circuit 38.

Logic circuit 38 also receives information from scan controller 30 regarding the timing of the scan cycle. This logic circuit 38 may be constructed to operate on the principles disclosed in the aforementioned British patent application. Alternatively, the logic circuit may include a memory device, and the memory may be compared with the results of sampling in one or multiple scans to determine whether the sampled output is correct and not noise. An example of a logic circuit that determines whether the black signal or white signal is correct and, if incorrect, replaces it with the correct one by interpolation, is the circuit shown in FIG.

The output of the logic circuit 38 is a series of pulses whose width varies in relation to the width of the scanned optical audio track, in which noise has been suppressed by the amplitude control and logic processing described above. . This pulse is integrated by an integrator or audio band filter 40 for an appropriate fixed period of time, and the resulting reproduced audio signal is applied to the theater's audio broadcasting system.

The embodiment shown in FIG. 4 is also applicable to systems having other types of scanning devices that require analog output signals. Devices of this type include, for example, those that scan to measure the thickness, width, hole diameter, etc. of a workpiece.

[Brief explanation of the drawing]

Fig. 1 is a graph showing an example of the relationship between density and distance when scanning a variable area optical audio track for understanding the present invention, and Fig. 2 is a graph showing an example of the relationship between density and distance when scanning a variable area optical audio track. 3 is a schematic perspective view showing in block form a portion of the optical reproduction device of the invention; FIG. 4 is a functional representation of the sensor array and processing circuitry in a preferred embodiment of the invention. 5 is a block diagram schematically showing a part of a logic circuit that can be used in FIG. 4, and FIG. 6 is a block diagram functionally showing a sensor array and a processing circuit in another embodiment of the present invention. It is a diagram. 10...Light source, 12...Slit, 14...Audio track, 16...Film, 18, 24...Sensor, 2
0... Processing circuit, 30... Scanning control device, 34... Amplifier, 36... Limiter, 37, 38... Logic circuit, 40
...Integrator, 48...Synthesizer.

Claims (1)

Claims: 1. A scanning device for reproducing a variable area optical soundtrack of a longitudinally moving motion picture film, comprising: a device for illuminating a substantial width of the soundtrack; and a device for illuminating a substantial width of the soundtrack; a plurality of light receiving devices disposed in a linear array laterally apart from the apparatus, each light receiving device generating an analog electrical signal in response to the level of illumination received from each minute portion of the soundtrack; a device for simultaneously and parallelly processing the electrical signals; and a device for simultaneously receiving and simultaneously combining the processed electrical signals to generate an audio signal. A scanning device having a device. 2. In the scanning device according to claim 1, the light receiving device is configured to randomly sample an uncertain region transitioning from a white area to a black area when the film moves in the longitudinal direction. A scanning device characterized in that the uncertainty region is converted into a dither signal in order to reduce the correlation between the electrical signal and the quantization noise by setting a sampling interval of . 3. A scanning device according to claim 1, in which said device processing in parallel further determines whether each signal is correct in response to said limited signals, and if incorrect, performs interpolation. A scanning device with a theoretical device that performs substitution by. 4. The scanning device according to claim 3, wherein the light receiving device is configured to randomly sample an uncertain region transitioning from a white area to a black area when the film moves in the longitudinal direction. A scanning device characterized in that the uncertainty region is converted into a dither signal in order to reduce the correlation between the electrical signal and the quantization noise by setting a sampling interval of . 5. The scanning device according to claim 2, wherein the sampling interval of the light receiving device is set so as to obtain two or three samples for each uncertain region. 6. A scanning device according to claim 4, characterized in that the sampling interval of the light receiving device is set so as to obtain two or three samples for each uncertain region.