JP4698617B2 - Image transmission apparatus and image transmission method - Google Patents

Description

  The present invention is a sensor that outputs a large-capacity image at a high rate with a limited wireless transmission speed, such as image transmission from an aircraft to the ground, and efficiently transmits a captured image to the ground in near real time according to the output rate. The present invention relates to an image transmission apparatus and an image transmission method.

Long-distance optical sensors and synthetic aperture radars mounted on airplanes have an enormous number of pixels per frame and an extremely high output rate. Therefore, generally, data is generally stored in a recorder, and after landing, the data is read from the recorder and analyzed on the ground, and it is impossible to instantaneously transmit a captured image from the aircraft to the ground.
In order to solve these problems, a high compression technique of image information such as JPEG or JPEG2000, and a method of thinning out image information or cutting out image information and transmitting to the ground are generally used.

For example, in an image processing method for displaying a recognition state of character recognition, a detailed recognition state of character recognition is displayed by displaying an evaluation value determined based on a correlation value indicating a similarity between a captured image and a template image. (For example, refer to Patent Document 1) and image processing on an image processing area in an image obtained by a camera based on a measurement result of a radar. A device that determines the center position of an image processing area and determines the size of the image processing area according to the beam profile of an electromagnetic wave output from a radar is known (see Patent Document 2).
Furthermore, there is a JPEG2000 tile encoding system as a high compression technology for image information, and ROI encoding technology already exists as another application example of tile division (Non-patent Document 1).

JP 2001-188904 A JP 2006-151125 A Ono Sadayasu and Suzuki Junji “Easy-to-Understand JPEG2000 Technology” Ohm, May 15, 2003, P80-82

When transmitting images to the ground by wireless transmission, image compression techniques such as JPEG are generally used, but large-capacity images of long-distance optical sensors and synthetic aperture radars targeted by the present invention are allowed by the Japanese radio law. When transmitting in near real time at a wireless communication speed, the information must be considerably degraded because it is compressed at a high compression rate, and the information must be transmitted with an image quality that is not suitable for detailed image analysis on the ground. There wasn't.
When an image is cut out, a necessary image range is cut out unnecessarily, and information necessary for detailed image analysis performed on the ground may be lost to the ground.
In addition, a method called template matching is generally used to automatically detect necessary image information from the entire area image. However, a large-capacity and high-speed frame rate sensor targeted by the present invention is used. Although it is possible to process captured images instantaneously if they are processed in parallel on a large-scale computer, etc., if it is necessary to install the device in a limited space such as an aircraft, installed weight, and power capacity, it is practically Impossible.

  Therefore, the present invention provides an image transmission apparatus and an image transmission method capable of transmitting a large-capacity and high-speed frame rate output image of a long-distance optical sensor or a synthetic aperture radar in near real time without degrading information. It is intended.

  In order to solve the above-described problems, the present invention provides an image transmission apparatus that transmits a captured image of a remote target to the ground, and the image is obtained from the target existence range obtained from the position evaluation result or the direction measurement result of the radio wave reverse detection apparatus. Detection range determination means for specifying a detection target range, template extraction means for specifying an image target from target identification candidates obtained from the radio wave reverse detection device, identification results of the detection range determination means and template extraction means, and a captured image Detailed target position calculating means for calculating the detailed target position from the above, adaptive compression parameter calculating means for calculating the optimum image compression parameter from the image output rate and the given wireless transmission speed, and the calculation result of the detailed target position calculating means And adaptive image compression means for performing image compression from the optimum image compression parameters of the adaptive compression parameter calculation means. The one in which the features.

Further, the present invention is a detection range determination step for calculating a region for performing image detection processing using the position evaluation result and the direction measurement result of the radio wave reverse detection device;
If the position rating result is obtained, calculating the target image frame and the detection target range in each image frame using the position rating result and its error range as inputs; and
When the position evaluation result is not obtained and only the square measurement result, calculating the detection target image frame and the detection target range in each image frame from the square measurement result and its square measurement error range;
Extracting a target template to be detected from identification candidates of a radio wave reverse detection device;
Performing image detection by template matching on the captured image from the detection target range obtained by the detection target range calculation step and the extracted template image group;
With respect to the detection result of the image detection step, a step of setting a pixel position having a correlation value equal to or higher than a preset threshold as a target position;
A step of setting a tile code size and a compression rate, which are compression parameters of JPEG 2000, taking into account a given transmission speed, an output rate and an image size of an image sensor, and a size and information amount of a target image area;
Performing JPEG2000 compression with the above compression parameters, adding detailed information for tiles in a region including the target image, and generating transmission image information;
An image transmission method comprising:

  The present invention performs the target position detection process using the result information of the radio wave reverse detection device, so that a large-capacity image of a long-distance optical sensor or a synthetic aperture radar can be obtained at a radio communication speed allowed by the Japanese radio law. Thus, it is possible to automatically transmit image information in near real time with an image quality that can withstand detailed image analysis performed on the ground. In particular, an image range including various targets such as a ship included in an image can be highly compressed using JPEG2000 encoding technology and transmitted to the ground so as not to deteriorate information.

Embodiment 1 FIG.
Hereinafter, specific embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a configuration of an image transmission apparatus according to Embodiment 1 of the present invention. In FIG. 1, 1 is an image sensor, 2 is a template database in which a plurality of template images are stored in advance, and 3 is a radio wave reverse detection device.
First, the captured image 4 is received from the image sensor 1 and input to the detailed target position calculation unit 7 and the adaptive image compression unit 11. Also, the target identification candidate 6 and the direction measurement result / position evaluation result 9 are received from the radio wave reverse detection device 3 and input to the template extraction means 5 and the detection range determination means 8. The template extracting means 5 is adapted to take in data from the template database 2 as appropriate.

  FIG. 2 shows a functional configuration diagram of the template extracting means 5 for selecting a template from the template database 2 and the target identification candidate 6. As shown in the figure, when there are a plurality of targets A to Z for the target identification candidates 6 sorted in the order of the identification accuracy, the threshold 16 is set in consideration of the processing load and speed, and the threshold 16 or more (in the figure, the target). A template data group 17 relating to the selected target is generated from the template database 2 with A to C) as detection targets.

Next, a processing example of the detection range determination unit 8 will be described with reference to FIGS. 3 and 4. FIG. 3 shows an example of processing when a position evaluation result is obtained from the radio wave reverse detection device 3, and is divided into tile areas having an aspect ratio of 4: 3, and each tile area is divided into, for example, times t = 1 to t = 6. The example which mapped each image frame 18 is shown.
In each image frame 18 from time t = 1 to t = 6, the position evaluation result range 21 is calculated in consideration of the position evaluation error 20 from the target position evaluation result 19. In each image frame 18 including the position evaluation result range 21, a detection range 22 (dotted line) for performing target detection processing is calculated by template matching. In FIG. 3, the image frame of t = 1 and t = 6 sets not only the entire area but only a part of the image as the detection range, t = 2 and t = 5 sets the entire area as the detection range, t = 3 and t = 4 are set as image frames outside the detection range.

  FIG. 4 is a processing example in the case where the position evaluation result is not obtained from the radio wave reverse detection device 3 and only the square measurement result is obtained. In each image frame 23 from time t = 1 to t = 9, the square measurement result range 25 is calculated in consideration of the square measurement error 25 from the direction measurement result and the azimuth line 24 obtained from the own position. In each image frame 23 including the square measurement result range 26, a detection range 27 (dotted line) for performing target detection processing is calculated by template matching. In FIG. 4, t = 1 and t = 3 image frames set only a part of the image as a detection range, and t = 2, t = 5, t = 4 and t = 9 image frames detect the entire region. Set with range. Further, t = 6, t = 7, and t = 8 are set as image frames outside the detection range.

  Next, a processing example of the detailed target position calculation means 7 will be described with reference to FIG. When the detailed target position calculation unit 7 receives an image frame including the detection range 27 calculated by the detection range determination unit 8 shown in FIG. The target search process by template matching is performed in the region within the detection range 27 for each image frame as shown in FIG. t = 1 and t = 3 target only a part of the image, and the target search process is performed on the entire region of the image frames of t = 2, t = 5, t = 4 and t = 9. Further, the target search process is not performed for t = 6, t = 7, and t = 8. As a result of the target search process, those having a correlation value by the template process equal to or higher than the threshold are determined as detection target positions and set as a detection result 10.

ここで、Ａは圧縮率、Ｂは無線伝送速度、Ｃは出力レート、ΔはJPEG2000の圧縮処理時間、目標検出処理時間および詳細画像タイル領域の情報量を考慮した調整係数である。Δは画像フレーム毎に動的に設定されるものとする。また、ＢおよびＣはbpsなどの同一単位に換算しておく必要がある。タイル符号サイズは、検出目標の画素数を十分含む大きさで設定するものとする。 Returning to the configuration block diagram of FIG. 1, the adaptive compression parameter calculation means 12 and the adaptive image compression means 11 will be described.
The adaptive compression parameter calculation means 12 calculates the compression rate used in JPEG 2000 and the tile code size when performing tile encoding from the given wireless transmission speed 13 and the output rate 14 of each image sensor 1. The compression rate is calculated by the following formula.

Here, A is a compression rate, B is a wireless transmission speed, C is an output rate, Δ is an adjustment coefficient that takes into account the JPEG 2000 compression processing time, target detection processing time, and the amount of information in the detailed image tile area. Δ is dynamically set for each image frame. B and C must be converted to the same unit such as bps. The tile code size is set to a size that sufficiently includes the detection target number of pixels.

  Next, a processing example of the adaptive image compression unit 11 will be described with reference to FIG. Since the target is not included in the image frame 30, only information that is JPEG 2000 compressed with the compression parameter calculated by the adaptive compression parameter calculation means 12 is transmitted as the transmission image information 15. On the other hand, since the target area 32 is included in the image frame 31, high-resolution information related to the detailed image tile range 33 is added in addition to the JPEG2000 compressed data equivalent to the image frame 31 and transmitted as the transmission image information 15. .

It is a block diagram which shows the structure of the image transmission apparatus which becomes Embodiment 1 of this invention. It is a function block diagram of the template extraction means 5 in FIG. It is explanatory drawing which shows the process example of the detection range determination means 8 from a position evaluation result. It is explanatory drawing which shows the process example of the detection range determination means 8 from a square measurement result. It is explanatory drawing which shows the process example of the detailed target position calculation means 7 in FIG. It is explanatory drawing which shows the process example of the adaptive image compression means 11 in FIG.

Explanation of symbols

1 image sensor, 2 template database,
3 Radio wave reverse detection device 4 Captured image
5 template extraction means, 6 target identification candidates,
7 detailed target position calculation means, 8 detection range determination means,
9 Square measurement result / Position evaluation result, 10 Detection result,
11 adaptive image compression means, 12 adaptive compression parameter calculation means,
13 wireless transmission speed, 14 output rate,
15 transmission image information, 16 threshold,
17 template data groups, 18 image frames,
19 Position rating result, 20 Position rating error range,
21 Position evaluation result range, 22 Detection range,
23 image frames, 24 bearing lines,
25 azimuth error range, 26 direction measurement range,
27 detection range, 28 extracted template data group,
29 template data, 30 image frames,
31 image frames, 32 target areas,
33 Detailed image tile range.

Claims (6)

In an image transmission device for transmitting a captured image of a remote target, detection range determination means for specifying a detection target range of an image from a target presence range obtained from a position evaluation result or a direction measurement result of a radio wave reverse detection device; A template extraction means for specifying an image target from target identification candidates obtained from the detection device, a detailed target position calculation means for calculating a detailed target position from the detection results and the identification results of the template extraction means and the captured image; An adaptive compression parameter calculation unit that calculates an optimal image compression parameter from an image output rate and a given wireless transmission speed, a calculation result of the detailed target position calculation unit, and an optimal image compression parameter of the adaptive compression parameter calculation unit An image transmission apparatus comprising adaptive image compression means for performing compression. 2. The image according to claim 1, wherein the detection range determination unit calculates a target image frame and a detection target range in each image frame by using a position evaluation result of the radio wave reverse detection device and an error range thereof as inputs. Transmission equipment. 2. The detection range determination unit according to claim 1, wherein the detection range determination unit calculates a detection target image frame and a detection target range in each image frame from a square measurement result of the radio wave reverse detection apparatus and a square measurement error range thereof. Image transmission device. 2. The template extraction unit according to claim 1, wherein when there are a plurality of identification candidates of the radio wave reverse detection device, a predetermined threshold value is set in advance, and all those above the threshold value are targeted. The image transmission apparatus described. 2. The image transmission apparatus according to claim 1, wherein the adaptive image compression unit performs image compression using a JPEG2000 tile encoding unit. A detection range determination step for calculating a region for performing image detection processing using the position evaluation result and the direction measurement result of the radio wave reverse detection device;
If the position rating result is obtained, calculating the target image frame and the detection target range in each image frame using the position rating result and its error range as inputs; and
When the position evaluation result is not obtained and only the square measurement result, calculating the detection target image frame and the detection target range in each image frame from the square measurement result and its square measurement error range;
Extracting a target template to be detected from identification candidates of a radio wave reverse detection device;
Performing image detection by template matching on the captured image from the detection target range obtained by the detection target range calculation step and the extracted template image group;
With respect to the detection result of the image detection step, a step of setting a pixel position having a correlation value equal to or higher than a preset threshold as a target position;
A step of setting a tile code size and a compression rate, which are compression parameters of JPEG 2000, taking into account a given transmission speed, an output rate and an image size of an image sensor, and a size and information amount of a target image area;
Performing JPEG2000 compression with the above compression parameters, adding detailed information for tiles in a region including the target image, and generating transmission image information;
An image transmission method comprising:

Images