会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 71. 发明专利
    • MOTION VECTOR DETECTING DEVICE
    • JPH04357773A
    • 1992-12-10
    • JP13121491
    • 1991-06-03
    • MITSUBISHI ELECTRIC CORP
    • TSUJI SHIGEKINISHIDA YOSHIHIRO
    • H04N5/14
    • PURPOSE:To obtain the accurate and simple motion vector of an image detecting device adopting the representative point matching system with short operating time. CONSTITUTION:The picture is divided into plural blocks, and pixel Rij of the representative point is set for each block. A vertical correlation detection part 11 (or horizontal correlation detection part 10) detects the block with high vertical correlation (or horizontal correlation) in the picture. A representative point memory 4 stores the pixel data for the representative point of the video signal for one frame. A horizontal picture element latch circuit 6 (or vertical picture element latch circuit 7) performs latching in the timing of the pixel Pij (x,0) (or Pij (0,y))corresponding to the horizontal direction (or vertical direction) of each representative point Rij. A horizontal vector detection part 8 (or vertical vector detection part 9) obtains the motion vector in the horizontal direction (or vertical direction) of the picture from the inter-frame correlation between the block Rij with high vertical correlation (or horizontal correlation) selected by the vertical correlation detection part 11(or horizontal correlation detection part 10) and the Pij (x,0) (or Pij (0,y)) in the next frame.
    • 72. 发明专利
    • MOTION VECTOR DETECTOR
    • JPH04310085A
    • 1992-11-02
    • JP7513991
    • 1991-04-08
    • MITSUBISHI ELECTRIC CORP
    • NISHIDA YOSHIHIROTSUJI SHIGEKI
    • H04N19/50H04N11/04H04N19/423H04N19/426H04N19/51H04N19/527H04N19/537H04N19/85
    • PURPOSE:To predict a motion vector with high accuracy by obtaining two one-dimensional vectors orthogonal to each other based on an accumulation table corresponding to a picture element at a point giving an offset of a predicted horizontal motion vector in the vertical direction and on an accumulation table corresponding to a relevant picture element in the horizontal direction. CONSTITUTION:An input video signal 2 is A/D-converted by an A/D converter 1 and written in a representing point memory 3. The stored data is read with a delay of one frame and fed to vertical and horizontal absolute circuits 9, 10 via a latch circuit 4. The video digital signal is fed to the vertical absolute value circuit 9 via a circuit 6 latched in the timing offset by (m) in the horizontal direction with respect to the timing of the picture element of the representative point in the vertical direction and fed to the horizontal absolute value circuit 10 via the circuit 8 latched in the timing offset by (n) in the vertical direction similarly. A difference between the signal of one preceding frame outputted from the circuit 4 and a signal from the circuit 6 is calculated by the circuit 9 and a motion vector is obtained from an output of the circuit 10 calculated similarly.
    • 73. 发明专利
    • CHARACTER/PATTERN STORING AND REPRODUCING DEVICE
    • JPH02262775A
    • 1990-10-25
    • JP8503889
    • 1989-04-03
    • MITSUBISHI ELECTRIC CORP
    • KIMURA KEIJINISHIDA YOSHIHIRO
    • H04N5/278H04N5/445
    • PURPOSE:To reproduce information as a character/pattern picture by providing a circuit to synthesize a signal, which is read from a memory, and an input signal during one cycle operation period of the memory, adding and storing new input information. CONSTITUTION:At the time of an add-on mode, an add-on signal is present and an enable signal (a) from a controller 11 is made H. For example, in the case of 8 bits, the former half of one cycle is read and the contents of a memory 6 in an address to be written are read and latched by a read data latch pulse E. In an OR circuit 17 of 8 bits, the contents are added with data, which are converted by a serial/parallel conversion pulse, and the latter half of one cycle is written to the memory 6. Thus, when the data are written to the memory in such a way, the input information of this time are added to information, which are stored up to the preceding time of operation, stored and go to the pattern/character information of the memory. Then, the information are finally added, superimposed with the other video input and go to a reproduced output.
    • 76. 发明专利
    • EDITING DEVICE
    • JPS63123285A
    • 1988-05-27
    • JP27067386
    • 1986-11-12
    • MITSUBISHI ELECTRIC CORP
    • KAWABATA MASARUNISHIDA YOSHIHIRO
    • H04N5/7826G11B27/02G11B27/032H04N5/782
    • PURPOSE:To simultaneously observe visually the connecting condition of videos after editing by storing the video signal of at least one field of the starting point and the completing point of respective editing sections, and projecting the result to a monitor displaying surface based on the editing sequence. CONSTITUTION:By a program key, a desired editing program is called out. Next, the operating key of a reproducing side VTR 2 is operated, a desired editing starting point is found and an inset key is inputted. A control part stores the counting value of a VTR 2 at this time and the reproducing video signal for one field at this time is stored to a video signal memory 32. Further, when a desired editing completing point is found, an outset key is inputted, the then tape counting value is stored and the reproducing video signal for one field at this time is stored to the memory 32. After the above-mentioned operation is repeated and the editing is completed, respective videos stored into the memory 32 are projected on the picture divided monitor TV4 in sequence of editing.
    • 77. 发明专利
    • MODE DISCRIMINATOR FOR MAGNETIC RECORDING AND REPRODUCING DEVICE
    • JPS60237658A
    • 1985-11-26
    • JP9495284
    • 1984-05-10
    • MITSUBISHI ELECTRIC CORP
    • NISHIDA YOSHIHIRO
    • G11B15/087G11B15/467
    • PURPOSE:To discriminate automatically the tape speed during recording by reproducing plural pilot signals of different frequencies by a rotary magnetic head, extracting and detecting the signal components for the frequency difference from the frequency of a carrier signal and comparing the same with a set threshold value. CONSTITUTION:The pilot signals of the frequencies f1-f4 are superposed on a video signal and are recorded in the diagonal tracks on a magnetic tape by a rotary magnetic head 8 in synchronization with the head switch signal to be inputted to a terminal 1. A selection switch 6 is set to the play side and the signal component for the frequency difference between the detected pilot signals and the frequency of the carrier signal is obtd. from a mixer 11 and is detected with BPFs 12, 13 in the stage of reproducing. The detected output is passed through a subtractor 16 and is made into the tracking error signal from a terminal 19. Said output is also supplied via gate circuits 30, 31 to comparators 21, 22 by which the output is compared with the set threshold value. A relational discrimination circuit 24 can discriminate the tape speed from the speed generator signal from a capstan motor inputted to a terminal 25 and the outputs from the comparators 21, 22.
    • 79. 发明专利
    • Method for generating pattern for automatic tracking of video tape recorder
    • 用于生成自动跟踪视频录像机的图形的方法
    • JPS59167825A
    • 1984-09-21
    • JP4333683
    • 1983-03-14
    • Mitsubishi Electric Corp
    • SHIROSHITA KENJIOOKAWA KAZUTAKANISHIDA YOSHIHIRO
    • H04N5/783G11B5/588G11B5/592
    • G11B5/5922
    • PURPOSE:To attain the generation of a head driving pattern with fidelity by preparing a tracking pattern and a transmission pattern as basic patterns in a pattern generator at the reproduction and combining an error pattern and the basic patterns so as to obtain a driving pattern. CONSTITUTION:When an output of an amplitude detector 14 is larger than an output of 14b at the 1/4 slow reproduction, it is meant that a video head 3a is biased toward a video track 2b on which a pilot signal f2 is recorded. Then, a moving head driving signal correcting the bias is generated at a pattern generating circuit 17 based on an output of a comparator 15 and a VTR operating mode signal 16 reprosenting the slow reproduction or the like. Further, the signal is smoothed by an LPF 18a smoothing it to a degree that an abnormal vibration is not produced to a piezoelectric element 4a and applied to the piezoelectric element 4a via a moving head driving amplifier 19a. As a result, the balance of the amount of crosstalk of a pilot signal from an adjacent track reproduced by the video head 3a is changed so as to form a servo loop, allowing to improve the tracking performance.
    • 目的:通过在再生中准备跟踪图案和作为图形发生器的基本图案的发送图案并组合误差图案和基本图案以获得驾驶模式,从而以保真度来产生头驱动模式。 构成:当在1/4慢再现时,幅度检测器14的输出大于14b的输出时,意味着视频头3a被偏压到其上记录有导频信号f2的视频轨道2b。 然后,基于比较器15的输出和代表慢速再现等的VTR操作模式信号16,在模式产生电路17产生校正偏置的移动头驱动信号。 此外,信号由LPF 18a平滑化,达到对压电元件4a不产生异常振动并经由移动头驱动放大器19a施加到压电元件4a的程度。 结果,改变了由视频头3a再现的相邻轨道的导频信号的串扰量的平衡,从而形成伺服环路,从而提高了跟踪性能。