U. S. ARMY. Technical Memorandum 9-65

Size: px

Start display at page:

Download "U. S. ARMY. Technical Memorandum 9-65"

Neal O’Neal’
5 years ago
Views:

1 U. S. ARMY Technical Memrandum 9-65 TARGET DETECTION USING BLACK-AND-WHITE TELEVISION STUDY I: THE EFFECTS OF RESOLUTION DEGRADATION ON TARGET DETECTION Lynn C. Oatman July 1965 AMCMSCde HUMAN ENGINEERING LABORATORIES ABERDEEN PROVING GROUND, MARYLAND C!. ' FOK v ' f /.CO <? r -^ ^ ^ Ct rlis I Distributin f this dcument is unlimited.

2 BLANK PAGES IN THIS DOCUMENT WERE NOT FILMED

AMCMS Cde 5011.11.841 Technical Memrandum 9-65 TARGET DETECTION USING BLACK-AND-WHITE TELEVISION STUDY I: THE EFFECTS OF RESOLUTION DEGRADATION ON TARGET DETECTION Lynn C.

3 AMCMS Cde Technical Memrandum 9-65 TARGET DETECTION USING BLACK-AND-WHITE TELEVISION STUDY I: THE EFFECTS OF RESOLUTION DEGRADATION ON TARGET DETECTION Lynn C. Oatman Technical Assistance William T. Nemeth Raymnd F. Blackmer July 1965 APPROV, )HN D. WEISZ Technical Directr Human Engineering Labratries U.S. ARMY HUMAN ENGINEERING LABORATORIES Aberdeen Prving Grund, Maryland Distributin f this dcument Is unlimited.

The fur levels f TV reslutin were presented t 16 subjects, wh were asked t indicate ir which ne f nine areas the tank appeared n the TV screen.

4 ABSTRACT The prbability f detectmg an M-48 tank at fur different levels f televisin (TV) reslutin (800, 600, 400, and 300 l^ies) was investigated n a black-and-white clsed-circuit TV system. The fur levels f TV reslutin were presented t 16 subjects, wh were asked t indicate ir which ne f nine areas the tank appeared n the TV screen. The data indicated that subjects perfrmed abut equally well at the 800, 600, and 400 levels f reslutin; hwever, their perfrmance was significantly prer at the 300 level f reslutiii. The tank's lcatin n the TV screen was an imprtant factr in the prbability f target detectin, but was cnfunded with ther variables. m

Percent Crrect Detectins as a Functin f Reslutin and Subjects. 9 3.

5 CONTENTS ABSTRACT iii INTRODUCTION 1 METHOD 3 Subjects 3 Apparatus 3 Prcedure 4 RESULTS 5 DISCUSSION 7 SUMMARY 8 REFERENCES 13 APPENDIX 15 FIGURES 1. Arrangement f Experimental Test and Cntrl Rms 2 2. Percent Crrect Detectins as a Functin f Reslutin and Subjects Mean Respnse Time as a Functin f Reslutin and Subjects Mean Respnse Time as a Functin f Reslutin and Lcatins Percent Crrect Detectins as a Functin f Lcatins 12

6 ' TABLES 1. Means and Standard Deviatins f Crrect Detectin Scres as a Functin f Reslutin 5 2. Means and Standard Deviatins f Respnse Times as a Functin f Reslutin 6 3. Summary f Analysis f Variance f Crrect Detectin Scres Summary f Analysis f Variance f Respnse Times 7 vi

' r- TARGET DETECTION USING BLACK-AND-WHITE TELEVISION STUDY I: THE EFFECTS OF RESOLUTION DEGRADATION ON TARGET DETECTION INTRODUCTION Televisin (TV) is becming an imprtant means f presenting visual

Fr example, TV has been used t present visual infrmatin t peratrs f re mte-cntrl cmbat vehicles (5). In anther study (6), men have driven tanks using a TV display as the sle surce f visual infrmatin.

7 ' r- TARGET DETECTION USING BLACK-AND-WHITE TELEVISION STUDY I: THE EFFECTS OF RESOLUTION DEGRADATION ON TARGET DETECTION INTRODUCTION Televisin (TV) is becming an imprtant means f presenting visual infrmatin t man in a variety f military applicatins. Televisin is currently being used fr many different military purpses. Fr example, TV has been used t present visual infrmatin t peratrs f re mte-cntrl cmbat vehicles (5). In anther study (6), men have driven tanks using a TV display as the sle surce f visual infrmatin. Televisin has als been used in lunar and planetary explratins and in satellites that prvide televised recnnaissance infrmatin. In military cntrl centers (2), TV displays are used fr mre effective weapn system deplyment and damage assessment. The increasing use f TV in mdern weapn and surveillance systems makes mre stringent demands n the peratrs' ability t accurately interpret TV displays. In sme military applicatins f TV, the peratr's task wuld be t detect and identify targets n the TV display. In a recent article, Grdn (4) has pinted ut that target detectin and target identificatin are tw imprtant aspects f the cmbat situatin. Target detectin rdinarily refers t an bserver's ability t determine whether a man-made bject is present r absent in a place where such bjects are nt usually seen. Target identificatin, n the ther hand, refers t the bserver's ability t characterize r specify the bject he has detected. Little attentin has been given t the extent t which TV reslutin affects the prbability f target dciectin. In a preliminary study, Oatman (7) cmpared targetlcatin scres frm a prjected-slide display and a TV display. Perfrmance was superir with the slide display, which was interpreted as due t better reslutin. Freeberg (3) fund that reducing infrmatin n a vide screen increased frm threshlds significantly. In a mre recent study, Shanahan (8) fund that reducing the bandwidth f a clsed-circuit TV system likewise reduced the subjects' ability t identify targets. The present study attempted t measure the effects f reslutin n the prbability f detecting targets n a black-and-white clsed-circuit TV system.

8 - A" K3 H- I a 3 CD 1 n 0) 3 s & A i O > < Q 3 ^ en 1 S? 1 CO 5' 'S s* 3 H ft) < ^3 3 M«rt O i P rt i N O 3

-601 clsed-circuit high-reslutin TV system. The TV camera in a cntrl rm fed radi-frequency (RF) signals t a 14-inch TV mnitr in an adjacent test rm.

9 METHOD Subjects Sixteen male enlisted men frm the U. S. Army served as subjects (Ss). The Ss' ages ranged frm 18 t 26. The Ss' visin (crrected r uncrrected) ranged frm 20/20 t 20/13 as measured by Orth-Rater tests. Apparatus Televisin The equipment fr the TV presentatin was a General Precisin Labratries P.D.-601 clsed-circuit high-reslutin TV system. The TV camera in a cntrl rm fed radi-frequency (RF) signals t a 14-inch TV mnitr in an adjacent test rm. A Ft-Vide F-101A light bx, hlding ne 8" by 10" psitivetransparency phtgraph, was placed in frnt f the TV camera. The TV mnitr in the test rm was divided int nine equal areas by black thread. The Ss had a respnse panel with nine buttns, which was similarly divided int nine areas by thin strips f masking tape. Each buttn n the respnse panel represented ne f the areas n the TV mnitr. The cntrl rm cntained a similar panel t display the Ss' respnses. The amunt f time that a picture remained n the TV mnitr was cntrlled by a Hunter interval timer, mdel 100-C, series D. A Standard electric timer recrded the Ss' respnse times. There was an intercm t prvide cmmunicatin between the test and cntrl rms. Figure 1 shws the experimental arrangements. Picture Degradatin Freeberg (3) pinted ut that bandwidth can be linearly related t the number f scan lines per frame. Hwever, in this experiment, the number f scan lines per frame in the actual displayed TV picture remained at 875 lines. The picture was degraded by an apprpriate reductin in bandwidth which, in turn, reduced the infrmatin in a hrizntal scan line. Thus the amunt f infrmatin within a scan line was decreased, but there was reductin nly in the hrizntal dimensin. T btain fur levels f picture degradatin (800, 600, 400, and 300, as measured with a standard RETMA TV test pattern), fur different values f capacitance t grund were inserted int the caxial cable between TV camera and mnitr. The capacitance t grund, in cnjunctin with ther elements

In the TV circuit, acted as a lw-pass filter. The bandwidth was reduced t 20.0, 6.0, 1.7, and 1 megacycles, which crrespnd t the 800, 600, 400, and 300 levels f reslutin respectively.

Prcedure After S was checked fr visual acuity, he was seated apprximately 72 inches frm the TV screen, instructed, and given the respnse panel.

10 In the TV circuit, acted as a lw-pass filter. The bandwidth was reduced t 20.0, 6.0, 1.7, and 1 megacycles, which crrespnd t the 800, 600, 400, and 300 levels f reslutin respectively. The amunt f picture degradatin (800, 600, 400, and 300) was measured with a standard RETMA TV test pattern, and these levels refer t reslutin in the center f the TV screen. Prcedure After S was checked fr visual acuity, he was seated apprximately 72 inches frm the TV screen, instructed, and given the respnse panel. Each S saw ten 8" x 10" phtgraphs, nine shwing an M-48 tank (ne in each area) against a hmgeneus backgrund, and ne shwing nly the backgrund. The tank was 3/8" lng and 3/16" wide, subtending a visual angle f 18 minutes. Every S saw the ten phtgraphs fur times, at each f the fur levels f picture degradatin. The phtgraphs were shwn in a randm rder, identical fr all Ss. T ffset practice effects, the fur levels f reslutin were presented in the same rder (800, 600, 400, and 300) t all Ss, but each S began with the level fllwing the ne at which the preceding S began. Fr example, the first S bserved the phtgraphs at the 800, 600, 400, and 300 levels f reslutin; the next S bserved them at the 600, 400, 300, and 800 levels; the third began with the 4Ü0 level; and s n, until all Ss had bserved the pntgraphs at all levels. Since the Ss were Initially selected at randm, assigning them t grups by this prcedure gave randm grups. The TV mnitr was always Illuminated. When the experimenter started the Interval timer, the picture was transmitted frm the cntrl rm t the Ss' mnitr In the test rm fr a 1.5-secnd expsure. If the S saw n target n the screen (and there was nne when the blank slide was presented), he pushed any tw buttns. v *. -^ <«*

RESULTS Tw aspects f the Ss' perfrmance were measured: respnse times and detectins. Respnse time was defined as the time between the target's appearance n the TV screen and the S's respnse.

11 RESULTS Tw aspects f the Ss' perfrmance were measured: respnse times and detectins. Respnse time was defined as the time between the target's appearance n the TV screen and the S's respnse. A detectin errr ccurred when S culd nt lcate the target n the TV screen. In scring, ne pint was allwed fr each crrect detectin. A cnstant value f ne was added t all treatment-by-subject cells in rder t avid blank cells in the analysis. This prcedure raised the means by ne but did nt affect the variance. Tables 1 and 2 present the means and standard deviatins fr the crrect detectin scres and respnse times, as a functin f picture degradatin. These means were analyzed with Duncan's (1) multiple-range test fr crrelated means. Detectin scres n the 300-level-reslutin TV display differed significantly ( <.01) frm thse n the 400-, 600-, and 800-level-reslutin TV displays. Hwever, the 400-, 600-, and 800-level-reslutin TV displays did nt differ significantly n detectin scres. With respnse time, nne f the reslutins differed significantly frm the thers. TABLE 1 Means and Standard Deviatins f Crrect Detectin Scres as a Functin f Reslutin (N = 144/cell) Reslutin Mean Standard Deviatin '4«M pw

TABLE 2 Means and Standard Deviatins f Respnse Times as a Functin f Reslutin (N = 144/cell) Reslutin S00 600 400 300 Mean 1.55 1.55 1.53 1.83 Standard Deviatin 0.51 0.76 0.47 0.

Hwever, the lcatins variable was the nly significant main effect (p <.01) fr respnse times. All f the interactins were significant (p <.

12 TABLE 2 Means and Standard Deviatins f Respnse Times as a Functin f Reslutin (N = 144/cell) Reslutin S Mean Standard Deviatin Tables 3 and 4 give summaries f the analyses f variance f detectin scres and respnse times. With crrect detectin scres, the main variables f reslutin and lcatins were significant (p <.01). Hwever, the lcatins variable was the nly significant main effect (p <.01) fr respnse times. All f the interactins were significant (p <.01) fr the respnse times; but nly tw interactins, reslutin by subjects and lcatins by subjects, were significant (p <.01) fr the detectin scres. TABLE 3 Summary f Analysis f Variance f Crrect Detectin Scres Surce df Sum f Squares Mean Square F Reslutins (A) ** Lcatins (B) ** Subjects (C) A x B AxC ** Bx C "* A x Bx C Ttal ** p <.01

TABLE 4 Summary f Analysis f Variance f Respnse Times Surce df Sum f Squares Mean Square F Reslutins (A) 3 10.31 3.43 2.43 Lcatins (B) 8 7.12 0.89 6.04** Subjects (C) 15 49.59 3.30 -- AxB 24 6.47 0.

13 TABLE 4 Summary f Analysis f Variance f Respnse Times Surce df Sum f Squares Mean Square F Reslutins (A) Lcatins (B) ** Subjects (C) AxB ** Ax C ** Bx C ** A x Bx C Ttal ** p <.01 DISCUSSION The results shw that reducing the bandwidth reduces target-detectin prbability significantly. Althugh it was expected that reductins in bandwidth and/r reslutin wuld affect the prbability f target detectin, it is interesting that the relatinship between reslutin and target detectin prbability is nnlinear. The data indicate a significant difference between 300 and 400; hwever, there were n significant differences between the 400, 600, and 800 levels f reslutin. One factr which may influence the btained relatinship is the fact that the reslutin was reduced nly in the hrizntal plane. It seems likely that the Ss' equivalent perfrmance n the 400, 600, and 800 levels f reslutin may be attributed t the unchanged vertical reslutin (650). Hwever, the Ss 1 perfrmance decreased at the lwest hrizntal reslutin level (300). The interactin f reslutin by subjects was significant fr bth detectin scres and respnse times. These interactins are pltted in Figures 2 and 3, bth f which shw nly tw levels (800 and 300) f reslutin. The figures shw that Ss perfrm quite differently at different levels f reslutin. With respnse time, there was a significant interactin between reslutin and lcatin. The plt f this interactin, in Figure 4, indicates that mean respnse time differs as a functin f reslutin and lcatin; hwever, fr the mst part, mean respnse time is faster with higher levels f reslutin.

The highest percentages f crrect detectins were in the upper middle, middle, and lwer middle areas f the TV screen. Percentage f crrect detectins was lwest in the crners f the TV screen.

14 The analysis f variance indicated that the target lcatin n the TV screen significantly affected the number f crrect detectins. The percentage f crrect detectins fr each f the nine areas n the TV screen is pltted in Figure 5. The highest percentages f crrect detectins were in the upper middle, middle, and lwer middle areas f the TV screen. Percentage f crrect detectins was lwest in the crners f the TV screen. Hwever, the effects f the target's lcatin n the TV screen are cnfunded by several ther factrs. First, Ss use different techniques t scan visual displays, s the btained lcatin effects culd ccur if the middle areas f the TV screen are scanned mre frequently than ther area'... Hwever, this scanning effect is cnfunded with a secnd factr: reslutin. With a bandwidth f 20 megacycles, reslutin is nt unifrm acrss the surface f the TV mnitr. When the center f the screen has 800-line reslutin, the level in the crner areas may be as lw as 500. Cnceivably, this accentuated degradatin in the TV display's crners culd accunt fr lw target-detectin prbabilities in the crners. A third factr which is cnfunded with the lcatin effect is the luminance f the TV screen. The TV screen's luminance, as measured with a Pritchard phtmeter (mdel 1970-PR), is nt unifrm. When the center area f the screen measures 7.6 ft-lamberts,- the crner areas may be as lw as 6.2 ft-lamberts. It is pssible, therefre, that these differing luminance values may als cntribute t lw target-detectin prbabilities in the crner areas. SUMMARY This experiment investigated hw TV reslutin affects target-detectin prbability. Sixteen Ss viewed an M-48 tank n a clsed-circuit black-and-white TV system under fur levels f reslutin (800, 600, 400, and 300). The Ss' task was indicating in which ne f nine areas the tank appeared. The data indicate that Ss' target-detectin prbability decreases significantly between the 300 and 400 levels f reslutin; hwever, there were n significant differences amng the 400, 600, and 800 levels f reslutin. In additin, the target's lcatin n the TV screen affected the target-detectin prbability significantly; hwever, this effect was cnfunded with reslutin, luminance, and Ss' search techniques.

$a. >* w > - > j\ 3" -r > -$ y. x i-l X v. V.

15 IVrccnt Crrect Detectins i^ a. >* w > - > j\ 3" -r > - y. x i-l X v. V. H C t > - 3 s n H c/5 ^i ) - cr r - -

16 01 _i Mean Respnse Time (sec.), to K> Ui C Oi I I L- c Oil I t.. ^ m. n X en ON«^J- c n H O ft) O pt Cß 00_ vo- -M 72 m en O S O -K z > a en PI n 51 N3 CO *.- X CO 00 en- -M O-J >

17 II Mean Respnse Time (sec.) -JL. -L- K3 b J d5-2 t fa w 3 U) ^ Ui M vo a CO C z n H z (0 rt 3 cn * en m r q a, z > I I OC 4^. O n > / / / / 00 m \0

$z\ t O I Percent Crrect$ - w H tfl O s z m > 4- - m -

18 z\ t O I Percent Crrect Detectins O -L JL - r S 8 U) - w H tfl O s z m > 4- - m - ^J * cn K) ^D ÜJ. 00 so a.

TT; r REFERENCES 1. Duncan, D. B. Multiple range tests fr crrelated and heterscedastic means. Bimetrics, 1957, 13, 164-176. 2. Electrnic Prving Grund.

19 TT; r REFERENCES 1. Duncan, D. B. Multiple range tests fr crrelated and heterscedastic means. Bimetrics, 1957, 13, Electrnic Prving Grund. Feasibility test f high-reslutin clsed-circuit televisin system. AD , U. S. Army, Frt Huachuca, Ariz., March Freeberg, N. E. Frm perceptin in vide viewing: Effjcts f reslutin degradatin and stere n frm threshlds. Technical Dcumentary Reprt ESD-TDR , Operatinal Applicatins Labratry, Bedfrd, Mass., December Grdn, D. A. Visual detectin and identificatin: Military applicatins. Memrandum R, Prject Michigan, University f Michigan, April Martindale, R. L. & Lwe, W. F. Use f televisin fr remte cntrl: A preliminary study. J. appl. Psychl., 1959, 43, Mler, C. G. & Brwn, G. L. Clsed-circuit televisin vehicle driving: I. A preliminary investigatin. Technical Memrandum 10-60, U. S. Army Human Engineering Labratries, Aberdeen Prving Grund, Md., August Oatman, L. C. Target detectin as a functin f expsure time and display mde. Technical Nte 8-63, U.S. Army Human Engineering Labratries, Aberdeen Prving Grund, Md., Octber Shanahan, D. Effects f televibin bandwidth n target identificatin. Technical Memrandum NCM-TM-64-2, U. S. Naval Missile Center, Pint Mugu, Calif., April

As yu see, it is in the grid n the screen. This panel is als marked int nine areas.

20 APPENDIX INSTRUCTIONS TO SUBJECTS We are ging t present a series f pictures n the screen. Each ne will be a phtgraph in which there may be a military target (tank), r the screen may ue blank. Here is an example (SLIDE ONE). Can yu spt the target? As yu see, it is in the grid n the screen. This panel is als marked int nine areas. As sn as yu spt the target, press the buttn that crrespnds with the area n the screen where yu see the target, Press the buttn as sn as pssible after yu spt the target. If the screen des nt have a target, push tw different buttns (any tw). I will say "Ready" befre I flash each slide. If the TV fails at any time, ntify me thrugh the intercm. D yu have any questins? 15

t Unclassified Security Classificatin DOCUMENT CONTROL DATA R&D (Smcurity claaalhcmlin l till», bdy l mbtltacl and indexing mnnlalin muaf be eniertd whan th» verall reprt ia c lastilied) ORIfilNATIN

2a REPORT SECURITY C L A»S I F I C A T I Of 2b CROUP Unclassified 3 REPORT TITLE TARGET DETECTION USING BLACK-AND-WHITE TELEVISION STUDY I: THE EFFECTS OF RESOLUTION DEGRADATION ON TARGET DETECTION 4

21 t Unclassified Security Classificatin DOCUMENT CONTROL DATA R&D (Smcurity claaalhcmlin l till», bdy l mbtltacl and indexing mnnlalin muaf be eniertd whan th» verall reprt ia c lastilied) ORIfilNATIN G ACTIVITY (Crprmla authr) U.S. Army Human Engineering Labratries Aberdeen Prving Grund, Md. 2a REPORT SECURITY C L A»S I F I C A T I Of 2b CROUP Unclassified 3 REPORT TITLE TARGET DETECTION USING BLACK-AND-WHITE TELEVISION STUDY I: THE EFFECTS OF RESOLUTION DEGRADATION ON TARGET DETECTION 4 DESCRIPTIVE NOTES (Typt l tprl and Incluaiva dataa) 5 AUJHOR(S) (Laal name, tiral nama, Initial) Oatman, Lynn C. 6 REPORT DATE July 1965 Ba CONTRACT OR GRANT NO 7«TOTAL NO OF PACES 20 9a ORIGINATOR'S REPORT NUMBERfS,) 7b NO OF REFS 8 b. PROJEC T NO Technical Memrandum b OTHER REPORT NO<S; (A n> ther numbers that may ba aaaignad thin reprt) to A VA (LABILITY LIMITATION NOTICES Distributin f this dcument is unlimited. Released t Dept. f Cmmerce fr sale t public. 11 SUPPLEMENTARY NOTES 12 SPONSORING MILITARY ACTIVITY 13 ABSTRACT The prbability f detecting an M-48 tank at fur different levels f televisin (TV) reslutin (800, 600, 400, and 300 lines) was investigated n a black-and-white clsedcircuit TV system. The fur levels f TV reslutin were presented t 16 subjects, wh were asked t indicate in which ne f nine areas the tank appeared n the TV screen. The data indicated that subjects perfrmed abut equally well at the 800, 600, and 400 levels f reslutin; hwever, their perfrmance was significantly prer at the 300 level f reslutin. The tank's lcatin n the TV screen was an imprtant factr in the prbability f target detectin, but was cnfunded with ther variables. DD.^ 1473 Unclassified Security Classificabn

TT Unclassified Security Classific.itin KtY *ORDS LINK A ROL E WT LINK B ROLE ft' LINK C ROLE WT Target Detectin Televisin Displays Human Engineering INSTRUCTIONS 1.

REPORT SECUHTY CLASSIFICATION: Enter the verall security classificatin f the reprt. Indicate whether "Restricted Data" is include^ Marking is t be in accrdance with apprpriate security regulutins. 2h.

22 TT Unclassified Security Classific.itin KtY *ORDS LINK A ROL E WT LINK B ROLE ft' LINK C ROLE WT Target Detectin Televisin Displays Human Engineering INSTRUCTIONS 1. ORIGINATING ACTIVITY: Enter the name and address f the cntractr, subcntractr, grantee. Department f Defense activity r ther rganizatin (crprate authr) issuing the reprt. 2a. REPORT SECUHTY CLASSIFICATION: Enter the verall security classificatin f the reprt. Indicate whether "Restricted Data" is include^ Marking is t be in accrdance with apprpriate security regulutins. 2h. GROUP: Autmatic dwngrading is specified in DD Directive and Armed Frces Industrial Manual. Enter the grup number. Als, when applicable, shw that ptinal markings have been used fr Grup 3 and Grup 4 as authrized. 3. REPORT TITLE: Enter the cmplete reprt tirle in all capital letters. Titles in all cases shuld be unclassified. If a meaningful title cannt be selected withut classificatin, shw title classificatin in all capitals in parenthesis immediately fllwing the title. 4. DESCRIPTIVE NOTES: If apprpriate, enter the type f reprt, e.g., interim, prgress, summary, annual, r final. Give the inclusive dates when a specific reprting perid is cvered. 5. AUTHOR(S): Enter the name(s) f authrs) as shwn n r in the reprt. Enter last name, first name, middle initial. If military, shw rank and branch f service. The name f the principal authr is an abslute minimum requirement. 6. REPORT DATE; Enter the date f Ihe reprt as day, mnth, year; r mnth, year. If mre than ne date appears n the reprt, use date f publicatin. 7a. TOTAL NUMBER OF PACFs The ttal page cunt shuld fllw nrmal paginatin prcedures, i.e., enter the number f pages cntaining infrmatin. 76. NUMBER OF REFERENCES; Enter the ttal number f references cited in the reprt. 8a. CONTRACT OR GRANT NUMBER: If apprpriate, enter the applicable number f the cntract r grant under which the reprt was written. 8b, 8c, 8t 8d. PROJECT NUMBER: Enter the apprpriate military department identificatin, such as prject number, subprject number, system numbers, task number, etc. 9a. ORIGINATOR'S REPORT NUMBER(S): Enter the fficial reprt number by which the dcument will be identified and cntrlled by the riginating activity. This number must be unique t this reprt. 9b. OTHER REPORT NUMBER(S): If the reprt has been assigned any ther reprt numbers (either by the riginatr r by the spnsr), als enter this number(s). 10. AVAILABILITY/LIMITATION NOTICES: Enter any limitatins n further disseminatin f the n prt, ther than thse impsed by security classificatin, using standard statements such as: (1) "Qualified requesters may btain cpies f this reprt frm DDC " (2) "Freign annuncement and disseminatin f this reprt by DDC is nt authrized." (3) "U. S. Gvernment agencies may btain cpies f this reprt directly frm DDC. Other qualified DDC users shall request 'trugh (4) "U. S. military agencies may btain cpies f this reprt directly frm DDC Other qualified users shall request thrugh (5) "All distributin f this reprt is cntrlled. Qualified DDC users shall request thrugh If the reprt has been furnished t the Office f Technical Services, Department f Cmmerce, fr sale t the public, indi cate this fact and enter the price, if knwn. 11. SUPPLEMENTARY NOTES; Use fr additinal explanatry ntes. 12. SPONSORING MILITARY ACTIVITY: Enter the name f the departmental prject ffice r labratry spnsring (pay mß fr) the research and develpment. Include address. 13. AFJSTRNCT; Enter an abstract giving a brief and factual summary f the dcument indicative f the reprt, even thugh it may als appear elsewhere in the bdy f the technical reprt. If additinal space is required, a cntinuatin sheet shall be attached. It is highly desirable that th.- abstract f classified re prts be unclassified. Each t aragraph f the abstract shall end with an indicatin f the military security classificatin f the infrmatin in the paragraph, represented as (TS), CS), (C), r (U). There is n limitatin n the length t the abstract. Hwever, Ihe suggested length is frm ISO t 225 wrds. 14. KEY WORDS: Key wrds are technically meaningful terms r shrt phrases that characterize a reprt and may be used as index entries fr catalging Ihe reprt. Key wrds must be selected s that n security classificatin is required. Idenfiers, such as equipment mdel designatin, trade name, nilitary prject cde name, gegraphic lcatin, may be used as key wrds b ' will be fllwed by an indicatin f technical cntext. The assignment f links, rules, and weights is ptinal Unclassified 3ec u ntyttassifica tin

CMV COMPETITION RULES

Atsusacn 2018 CMV COMPETITION RULES Atsusacn 2018 CMV cmpetitin rules Pagina 1 GENERAL RULES The cntent must be apprpriate fr ages 13 and lder. 1 70% f the used images must cntain csplays base dn (Japanese)